diff --git a/data/adapter_stats.json b/data/adapter_stats.json index 0da3c8aca4b460a56ec6f63ee4b126db13a986a6..7113eefbd2ad0bae939ccae9679ba57b880b3048 100644 --- a/data/adapter_stats.json +++ b/data/adapter_stats.json @@ -30,43 +30,71 @@ "last_failed": true }, "global-mmlu-lite": { - "data_fingerprint": "5ef9a59574c1b93832c2f2242d9b3258ea081dae80fdf6157c416951eaecc313", + "data_fingerprint": "728e4a14097f3a2aa6266537bfe36ed95167f86c0efb8753e537ba1982abe49d", "entry_hashes": [ + "6b6f6f4959ccd5ca65d63f5061c5681c3aa8e1492676a130ff392eaf80814c6e", "4651ceff3c95784f330c369c71b6e005369f9d1dc716c6e499d336f3d1a85e14", - "2f6ebe713380721bc675985a219b89f3f6e5d3cb1ac0406bb164950d171e49da", - "617541926325b2677abead9ff73c17b3ea8c4033d3c54a9dbacfcdd2a1675881", + "f00df60c831026625621911d570f47fc98a42f82d38cb2a0a2a285e8f8ba870b", + "8e82dcad462992c82b749ea0bd9babb8fddd02ed046775bfc1e4d13af0838216", "f0ccfb7d0ca71985c8997a8a977b3c9c299627c13107639240350a4715c75a0e", - "9e78a0d77dc1d35918738b48c17165a29d24b90170c0375e636aae29c1b46e08", - "2bb2af5e0b5152318390230587a97fe6ac3ebf447d4fab5784308d5ff0fb0804", - "0899b9c965496891cbb846b65d0f68c152a81c2b96700941557400d344090da7", - "87c2ec8681f72781d56acdd3e0c9fcd6fa84ac462d166d5d78bddf1e4343b590", - "7d6adaf2431ebaf7f05da4619b32ebde41eb026aea16d3bd8aecec98ac464bdc", - "2895034784c6c5642892b7c992c973047f2146097c6a13dc155f292a5d52867e", - "b18f637032ba10884f65a463b7b28860a671187534bd27097fed2c3041990c28", - "a0dae39f0c46d5c2d066e2516193d212c4952ded26e7767521ed53df91dcb5e9", - "0a6682601e86bd48c3248771f21c03db15ca13fca4c2bb4b52d0abb5bb37340e", + "13b79500d0bcf8426174175d06a892189da9cf587a3884d253ca3edc6b611150", + "aa570ef42f0c5477de874699226ddee44c86dde134f440e2f3ed5201bc3a59f5", "a55466285b5942f464cd00ec1ef0925e2f7b6a561005b2aad88be014ed62dc11", - "f00df60c831026625621911d570f47fc98a42f82d38cb2a0a2a285e8f8ba870b", - "b47cc80c245540e9e01ce8fd850cc8e071ddb4db644671c4a7b23edd487f93ba", + "2895034784c6c5642892b7c992c973047f2146097c6a13dc155f292a5d52867e", "cf1434a0eba892b55cd79729aa53923057d629e6d1b264117cce480d172c79e8", - "fee4a524d656eb05c8282f5c9762680946785507fa019968843997cafdf367a0", "e6fc3bcd91191aeebfcfaa4b52eb256c3759c6d06f98674af0fea97dc9fe8ced", - "a757485a658f13df5452e8a5bb2534332039faaf8c5e0f34c7167ab446b94ae7", - "8e82dcad462992c82b749ea0bd9babb8fddd02ed046775bfc1e4d13af0838216", + "97a83d95e9f5ed82780c3492a02666d8b05c70d7d1641aeb0a750bcd91cd82c5", + "e1654549991d6bd2be1243a7f9d4742c24b80cc6155b78c87b44034989804711", + "87c2ec8681f72781d56acdd3e0c9fcd6fa84ac462d166d5d78bddf1e4343b590", + "fee4a524d656eb05c8282f5c9762680946785507fa019968843997cafdf367a0", + "58a9227f563922acc76feee11289f1d6a0e7244f4a9f8194f613a0c12b1ac5b1", + "182d7607b74a3e07eb9237cd73e64bc7e6521a198b1063d1cc532786c801712a", + "94c6b8ccb15f6bdb2bcc32b2b6ab405b36157d4eff55680248489c082d8be3cc", + "56baf3d5bb3c762fde458787048f09e461e99ad925828c57a98a32ff9ea064af", + "3afa76ca8cf7122cdda457ef422b1df24025b292bf81dd157f091027e369c233", + "b47cc80c245540e9e01ce8fd850cc8e071ddb4db644671c4a7b23edd487f93ba", + "a0dae39f0c46d5c2d066e2516193d212c4952ded26e7767521ed53df91dcb5e9", + "7ef02ad4eba043ca76f905b2fd6e0c46502c19a1658224240c60828640ca3a83", + "617541926325b2677abead9ff73c17b3ea8c4033d3c54a9dbacfcdd2a1675881", + "95907b7006252515d0333b6819f25a9ca3c1527d8e9d9c79c4f2dfd8dc00cdd8", + "0899b9c965496891cbb846b65d0f68c152a81c2b96700941557400d344090da7", + "1f96571a09759746d382363de75db25816df1bcb6cbf3ec552b39854cad0b67f", + "9bbbcfc11fa3bf13eaed47af2835e72c5a143027da7d564f719eb0952e0008c1", + "039d51e4e23fe38ea363b9b4dbbb26ae227d59ae3c5cd5d40629ba0d4eceeb17", + "0a6682601e86bd48c3248771f21c03db15ca13fca4c2bb4b52d0abb5bb37340e", + "5b4feebced3af0e9d6a28fc1de150b82555fdeeaffbe8285f98d759c5f61dae7", + "7d6adaf2431ebaf7f05da4619b32ebde41eb026aea16d3bd8aecec98ac464bdc", + "c46c83cec65ac806a031b9ecf8c647d78f3630cea6472a101e3e22ea4b8b766c", + "bd5a0e428f30a98e50e4da8b130387c744235458a1d069b609b8e1877048d6f1", "ec4833f773abcb0d0b375aebbb35cb2846595fd5064b69121011748ae847b211", - "aa570ef42f0c5477de874699226ddee44c86dde134f440e2f3ed5201bc3a59f5", + "a7b686a4a8594668bd17e5e9cb52fa7a5c93f930ba4569d696bc71c3d4521ed3", + "9e78a0d77dc1d35918738b48c17165a29d24b90170c0375e636aae29c1b46e08", + "023eb75410dc101e5369ec8cf3fdeb12a37b865ca3bacfbfaa319bc4ff537c9d", + "b18f637032ba10884f65a463b7b28860a671187534bd27097fed2c3041990c28", + "2f6ebe713380721bc675985a219b89f3f6e5d3cb1ac0406bb164950d171e49da", "b1cb0e2c400ce58b6c9dbf2eb67e625fa4a838075a78f01eabac686221c0ba32", - "3afa76ca8cf7122cdda457ef422b1df24025b292bf81dd157f091027e369c233", - "a8853193324cd933332f1e4ee589217e6b756dadc364e81b200410544fc571d8", + "7267510a3e0c1b005475ca19137c440c1b5e3b4039c8da55457b1cecad45586d", + "b0a82fa63ae27f8eb6c609bc0aed5ad807dcab35eb09e2468494ed03e137ec0c", + "b79ec1c965d1b92146318c54e61a642afb7a74e1410ad08c82856a17c738118e", + "a757485a658f13df5452e8a5bb2534332039faaf8c5e0f34c7167ab446b94ae7", + "1c0598d106085627d47f071c942f8202b6df32a95ba4ae174bc4e42b083d3ea6", + "a6aab998f015ea0cacae438b9a78f42d90c8057c39ca28bb09e3cb03d396645e", + "58210e7d713543a1244c08d778845736a8b36523602b55f0eab5a9b4fbcba402", "0783abf527a9d3f8165081d8cb2e33758e68b066ba18eca41c7d97ff06ee2ca9", - "58a9227f563922acc76feee11289f1d6a0e7244f4a9f8194f613a0c12b1ac5b1" + "a6568d4775433a5de47b14029fa8ec2301eac9b067fdcf241120a9223d1a42fd", + "2bb2af5e0b5152318390230587a97fe6ac3ebf447d4fab5784308d5ff0fb0804", + "d506cb354b00b6039fc00a5343f5b40290ae4793f7f76fcc498afd27894955d0", + "e9fe43ba51dfdf442d3810e3b8dbe62f6333fc2b6c569dd9aead6f400b7f89b7", + "a8853193324cd933332f1e4ee589217e6b756dadc364e81b200410544fc571d8", + "cc11505004f63ce30c457bcf5efef7312cf4a8c8342ab62e4ec1d48c88fefc25", + "940995c822dcd7d8982bad6fd3e80fdbb65d9fbcb75c5d5b1157f52967cd8c67" ], - "last_check_ts": 1783601731.979473, - "last_data_change_ts": 1783518896.607522, + "last_check_ts": 1783820828.8573594, + "last_data_change_ts": 1783820828.8573608, "last_failed": false, - "last_success_ts": 1783601731.9794726, - "size_mb": 0, - "time_s": 1.4341261386871338 + "last_success_ts": 1783820828.8573592, + "size_mb": 0.3680877685546875, + "time_s": 1.479100227355957 }, "hal": { "data_fingerprint": "7bb18f848bce3afee9bf1be8268e2dc115892db8dc93bd8d3daa5cbf21b60ead", @@ -4992,7 +5020,7 @@ "url_etag": "W/80766304581529293" }, "hle": { - "data_fingerprint": "2271963061e62faf7ad57d53687bd26acdc0a5c4824f567d0344498524270633", + "data_fingerprint": "", "entry_hashes": [ "368237761d719dda2f682038eac58a17e6150d09e77db1054eab8813ecb5a2fb", "b6ca2fc5888e85a520d37ad92a922794cc1d2a4daf6989d8da6ced059b5d41f3", @@ -5045,12 +5073,12 @@ "2d14bf88d84c7b222594b1c3d96605420459a412ad57eb4f01506d73db8e195f", "651909ce68911a0049d12fdfd510758a1603338dc60f80a967746a2046b37813" ], - "last_check_ts": 1783601731.8805106, + "last_check_ts": 1783820828.8215966, "last_data_change_ts": 1783518897.3675156, "last_failed": false, - "last_success_ts": 1783601731.8805096, + "last_success_ts": 1783820828.821596, "size_mb": 0, - "time_s": 1.5643982887268066 + "time_s": 1.448526382446289 }, "livecodebenchpro": { "last_failed": true @@ -5538,7 +5566,7 @@ "time_s": 2.330172300338745 }, "multi_swe_bench": { - "data_fingerprint": "89a0196eeabde307083ea8312416acf3aa63dd9ed391ef39dbbb2f0d1371f3a6", + "data_fingerprint": "", "entry_hashes": [ "f753b7c97aba6b2d8dc03cd7bbd02b8fab2f73d65930442bbe51e984368257c7", "a39aee5acb478d354c0844ef682430f74f9ad42c902d3fa25bae02cd21e697a9", @@ -5803,15 +5831,15 @@ "a401a3c328a28c918f2fa2d30bc331d2cf739ca5580dd662a5a4e2e12d5abc8e", "500a51d34b43447f7354d7c055f958fb02e8f0893529031d802f3c65b80aef42" ], - "last_check_ts": 1783601763.9393828, + "last_check_ts": 1783820857.880977, "last_data_change_ts": 1783518930.3473694, "last_failed": false, - "last_success_ts": 1783601763.9393826, + "last_success_ts": 1783820857.8809764, "size_mb": 0, - "time_s": 33.76141691207886 + "time_s": 30.83329677581787 }, "openeval": { - "data_fingerprint": "4eba60df14d811b62d1f3b2d77cc2b26e2a6773bc051e2603f5ee8cc8f8315e6", + "data_fingerprint": "", "entry_hashes": [ "6266db2b16a676e3a62eec7195be23f576929bafc1dd8471841609c83f9106e7", "550063401f9d81b1e42ba0d5bb04456c5872e2c31c58fd164609e92860f7e36c", @@ -6926,12 +6954,12 @@ "592d23a34610414f0f540a25831f981a19790f0fe4434bab39715860fde64f42", "6f74e246ae81a8de415c23c5798410c22aaadaddbc9483cb8296de7bf18b2b45" ], - "last_check_ts": 1783602516.2646487, + "last_check_ts": 1783821575.2712789, "last_data_change_ts": 1783519731.9760447, "last_failed": false, - "last_success_ts": 1783602516.2646482, + "last_success_ts": 1783821575.2712786, "size_mb": 0, - "time_s": 785.5236105918884 + "time_s": 747.5695848464966 }, "rewardbench": { "last_failed": true, diff --git a/data/global-mmlu-lite/alibaba/qwen3-235b-a22b-instruct-2507/9b88ab5b-4963-4ab9-a381-b2ed8d4c17b5.json b/data/global-mmlu-lite/alibaba/qwen3-235b-a22b-instruct-2507/9b88ab5b-4963-4ab9-a381-b2ed8d4c17b5.json new file mode 100644 index 0000000000000000000000000000000000000000..8f49fd2fb35d61b2efeae0a68a6e4b7a89076a8d --- /dev/null +++ b/data/global-mmlu-lite/alibaba/qwen3-235b-a22b-instruct-2507/9b88ab5b-4963-4ab9-a381-b2ed8d4c17b5.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/alibaba_qwen3-235b-a22b-instruct-2507/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "qwen3-235b-a22b-instruct-2507", + "id": "alibaba/qwen3-235b-a22b-instruct-2507", + "developer": "alibaba", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Qwen 3 235B A22B Instruct 2506" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8798 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8522 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.89, + "uncertainty": { + "confidence_interval": { + "lower": -0.0307, + "upper": 0.0307, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885, + "uncertainty": { + "confidence_interval": { + "lower": -0.0313, + "upper": 0.0313, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0321, + "upper": 0.0321, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0321, + "upper": 0.0321, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/anthropic/claude-3-5-haiku-20241022/d180f6bf-a455-4741-915f-cd3105b194ac.json b/data/global-mmlu-lite/anthropic/claude-3-5-haiku-20241022/d180f6bf-a455-4741-915f-cd3105b194ac.json new file mode 100644 index 0000000000000000000000000000000000000000..9991d05d539809b9bd66b1d017f968dfd71aa158 --- /dev/null +++ b/data/global-mmlu-lite/anthropic/claude-3-5-haiku-20241022/d180f6bf-a455-4741-915f-cd3105b194ac.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/anthropic_claude-3-5-haiku-20241022/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "claude-3-5-haiku-20241022", + "id": "anthropic/claude-3-5-haiku-20241022", + "developer": "anthropic", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Claude 3.5 Haiku" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6114 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5834 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6394 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.695, + "uncertainty": { + "confidence_interval": { + "lower": -0.0451, + "upper": 0.0451, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.485, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0459, + "upper": 0.0459, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.565, + "uncertainty": { + "confidence_interval": { + "lower": -0.0486, + "upper": 0.0486, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.61, + "uncertainty": { + "confidence_interval": { + "lower": -0.0478, + "upper": 0.0478, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0488, + "upper": 0.0488, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.48, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.655, + "uncertainty": { + "confidence_interval": { + "lower": -0.0466, + "upper": 0.0466, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5225, + "uncertainty": { + "confidence_interval": { + "lower": -0.0489, + "upper": 0.0489, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.485, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.69, + "uncertainty": { + "confidence_interval": { + "lower": -0.0453, + "upper": 0.0453, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0462, + "upper": 0.0462, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.69, + "uncertainty": { + "confidence_interval": { + "lower": -0.0453, + "upper": 0.0453, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7, + "uncertainty": { + "confidence_interval": { + "lower": -0.0449, + "upper": 0.0449, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/anthropic/claude-3-7-sonnet-20250219/2f5ce580-87fa-4e08-85e1-958db01a560b.json b/data/global-mmlu-lite/anthropic/claude-3-7-sonnet-20250219/2f5ce580-87fa-4e08-85e1-958db01a560b.json new file mode 100644 index 0000000000000000000000000000000000000000..f8ba079f483f21a59524b62aa9cf12f6151fd1b1 --- /dev/null +++ b/data/global-mmlu-lite/anthropic/claude-3-7-sonnet-20250219/2f5ce580-87fa-4e08-85e1-958db01a560b.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/anthropic_claude-3-7-sonnet-20250219/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "claude-3-7-sonnet-20250219", + "id": "anthropic/claude-3-7-sonnet-20250219", + "developer": "anthropic", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Claude 3.7 Sonnet" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8078 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7794 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8362 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7625, + "uncertainty": { + "confidence_interval": { + "lower": -0.0417, + "upper": 0.0417, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0372, + "upper": 0.0372, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0414, + "upper": 0.0414, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.805, + "uncertainty": { + "confidence_interval": { + "lower": -0.0388, + "upper": 0.0388, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8175, + "uncertainty": { + "confidence_interval": { + "lower": -0.0379, + "upper": 0.0379, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8225, + "uncertainty": { + "confidence_interval": { + "lower": -0.0374, + "upper": 0.0374, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0357, + "upper": 0.0357, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.83, + "uncertainty": { + "confidence_interval": { + "lower": -0.0368, + "upper": 0.0368, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.77, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0386, + "upper": 0.0386, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.81, + "uncertainty": { + "confidence_interval": { + "lower": -0.0384, + "upper": 0.0384, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.835, + "uncertainty": { + "confidence_interval": { + "lower": -0.0364, + "upper": 0.0364, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/anthropic/claude-opus-4-1-20250805/dff6ca0e-03eb-4fb6-8096-5ccfeb5b66ae.json b/data/global-mmlu-lite/anthropic/claude-opus-4-1-20250805/dff6ca0e-03eb-4fb6-8096-5ccfeb5b66ae.json new file mode 100644 index 0000000000000000000000000000000000000000..f1745936cbbe2864a7b829a5592976ff0d36ee86 --- /dev/null +++ b/data/global-mmlu-lite/anthropic/claude-opus-4-1-20250805/dff6ca0e-03eb-4fb6-8096-5ccfeb5b66ae.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/anthropic_claude-opus-4-1-20250805/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "claude-opus-4-1-20250805", + "id": "anthropic/claude-opus-4-1-20250805", + "developer": "anthropic", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Claude Opus 4.1" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.943 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9331 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9528 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.93, + "uncertainty": { + "confidence_interval": { + "lower": -0.025, + "upper": 0.025, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0237, + "upper": 0.0237, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/anthropic/claude-opus-4-8-default/e18bd0fb-8f2c-4016-aafb-6ff1e6fc5a55.json b/data/global-mmlu-lite/anthropic/claude-opus-4-8-default/e18bd0fb-8f2c-4016-aafb-6ff1e6fc5a55.json new file mode 100644 index 0000000000000000000000000000000000000000..b0cc51cdb7f1f77d3865d6be733ac16b28c0a619 --- /dev/null +++ b/data/global-mmlu-lite/anthropic/claude-opus-4-8-default/e18bd0fb-8f2c-4016-aafb-6ff1e6fc5a55.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/anthropic_claude-opus-4-8-default/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "claude-opus-4-8-default", + "id": "anthropic/claude-opus-4-8-default", + "developer": "anthropic", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Claude Opus 4.8" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9485 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9387 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9584 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9573, + "uncertainty": { + "confidence_interval": { + "lower": -0.0199, + "upper": 0.0199, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0198, + "upper": 0.0198, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0208, + "upper": 0.0208, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9494, + "uncertainty": { + "confidence_interval": { + "lower": -0.0216, + "upper": 0.0216, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/anthropic/claude-sonnet-4-20250514/e2bf6d75-370d-47e0-b027-8b1bd88b4190.json b/data/global-mmlu-lite/anthropic/claude-sonnet-4-20250514/e2bf6d75-370d-47e0-b027-8b1bd88b4190.json new file mode 100644 index 0000000000000000000000000000000000000000..c25cbb1c099532f361fcad21bf718d737ab68674 --- /dev/null +++ b/data/global-mmlu-lite/anthropic/claude-sonnet-4-20250514/e2bf6d75-370d-47e0-b027-8b1bd88b4190.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/anthropic_claude-sonnet-4-20250514/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "claude-sonnet-4-20250514", + "id": "anthropic/claude-sonnet-4-20250514", + "developer": "anthropic", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Claude Sonnet 4" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9058 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8913 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9203 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9, + "uncertainty": { + "confidence_interval": { + "lower": -0.0294, + "upper": 0.0294, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9, + "uncertainty": { + "confidence_interval": { + "lower": -0.0294, + "upper": 0.0294, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0297, + "upper": 0.0297, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0297, + "upper": 0.0297, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9175, + "uncertainty": { + "confidence_interval": { + "lower": -0.027, + "upper": 0.027, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0304, + "upper": 0.0304, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/cohere/aya-expanse-32b/c8256802-cef0-441a-8188-b4fd9a447893.json b/data/global-mmlu-lite/cohere/aya-expanse-32b/c8256802-cef0-441a-8188-b4fd9a447893.json new file mode 100644 index 0000000000000000000000000000000000000000..daaa2bb48f477c83538cb0436933b40c463082dd --- /dev/null +++ b/data/global-mmlu-lite/cohere/aya-expanse-32b/c8256802-cef0-441a-8188-b4fd9a447893.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/cohere_aya-expanse-32b/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "aya-expanse-32b", + "id": "cohere/aya-expanse-32b", + "developer": "cohere", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Aya Expanse 32B" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7353 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6891 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7815 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7544, + "uncertainty": { + "confidence_interval": { + "lower": -0.0422, + "upper": 0.0422, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7343, + "uncertainty": { + "confidence_interval": { + "lower": -0.0433, + "upper": 0.0433, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7325, + "uncertainty": { + "confidence_interval": { + "lower": -0.0434, + "upper": 0.0434, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0431, + "upper": 0.0431, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7594, + "uncertainty": { + "confidence_interval": { + "lower": -0.0419, + "upper": 0.0419, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7305, + "uncertainty": { + "confidence_interval": { + "lower": -0.0436, + "upper": 0.0436, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7419, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7544, + "uncertainty": { + "confidence_interval": { + "lower": -0.0422, + "upper": 0.0422, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7362, + "uncertainty": { + "confidence_interval": { + "lower": -0.0433, + "upper": 0.0433, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7071, + "uncertainty": { + "confidence_interval": { + "lower": -0.0448, + "upper": 0.0448, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6942, + "uncertainty": { + "confidence_interval": { + "lower": -0.0452, + "upper": 0.0452, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.743, + "uncertainty": { + "confidence_interval": { + "lower": -0.0432, + "upper": 0.0432, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0448, + "upper": 0.0448, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/cohere/command-a-03-2025/dc57aed9-944f-4d35-b965-9a800afe812a.json b/data/global-mmlu-lite/cohere/command-a-03-2025/dc57aed9-944f-4d35-b965-9a800afe812a.json new file mode 100644 index 0000000000000000000000000000000000000000..2a667fac6022b9ee42d9941196e2f452575ba9d1 --- /dev/null +++ b/data/global-mmlu-lite/cohere/command-a-03-2025/dc57aed9-944f-4d35-b965-9a800afe812a.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/cohere_command-a-03-2025/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "command-a-03-2025", + "id": "cohere/command-a-03-2025", + "developer": "cohere", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Command A " + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8385 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7993 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8778 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0357, + "upper": 0.0357, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.855, + "uncertainty": { + "confidence_interval": { + "lower": -0.0345, + "upper": 0.0345, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8225, + "uncertainty": { + "confidence_interval": { + "lower": -0.0374, + "upper": 0.0374, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0357, + "upper": 0.0357, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0362, + "upper": 0.0362, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8421, + "uncertainty": { + "confidence_interval": { + "lower": -0.0358, + "upper": 0.0358, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8546, + "uncertainty": { + "confidence_interval": { + "lower": -0.0346, + "upper": 0.0346, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0362, + "upper": 0.0362, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.845, + "uncertainty": { + "confidence_interval": { + "lower": -0.0355, + "upper": 0.0355, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.85, + "uncertainty": { + "confidence_interval": { + "lower": -0.035, + "upper": 0.035, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.84, + "uncertainty": { + "confidence_interval": { + "lower": -0.0359, + "upper": 0.0359, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0348, + "upper": 0.0348, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8275, + "uncertainty": { + "confidence_interval": { + "lower": -0.037, + "upper": 0.037, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.815, + "uncertainty": { + "confidence_interval": { + "lower": -0.0381, + "upper": 0.0381, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.835, + "uncertainty": { + "confidence_interval": { + "lower": -0.0364, + "upper": 0.0364, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8175, + "uncertainty": { + "confidence_interval": { + "lower": -0.0379, + "upper": 0.0379, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/deepseek/deepseek-r1-0528/065c6aa2-454e-4ced-acc9-97bd5d492239.json b/data/global-mmlu-lite/deepseek/deepseek-r1-0528/065c6aa2-454e-4ced-acc9-97bd5d492239.json new file mode 100644 index 0000000000000000000000000000000000000000..f5ab22ca2ac59ea79f071662195d238b7b720d26 --- /dev/null +++ b/data/global-mmlu-lite/deepseek/deepseek-r1-0528/065c6aa2-454e-4ced-acc9-97bd5d492239.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/deepseek_deepseek-r1-0528/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "deepseek-r1-0528", + "id": "deepseek/deepseek-r1-0528", + "developer": "deepseek", + "inference_platform": "unknown", + "additional_details": { + "display_name": "DeepSeek-R1" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6744 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6672 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6816 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0456, + "upper": 0.0456, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.715, + "uncertainty": { + "confidence_interval": { + "lower": -0.0442, + "upper": 0.0442, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.655, + "uncertainty": { + "confidence_interval": { + "lower": -0.0466, + "upper": 0.0466, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0471, + "upper": 0.0471, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0452, + "upper": 0.0452, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0468, + "upper": 0.0468, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.655, + "uncertainty": { + "confidence_interval": { + "lower": -0.0466, + "upper": 0.0466, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0458, + "upper": 0.0458, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0411, + "upper": 0.0411, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.635, + "uncertainty": { + "confidence_interval": { + "lower": -0.0472, + "upper": 0.0472, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7175, + "uncertainty": { + "confidence_interval": { + "lower": -0.0441, + "upper": 0.0441, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0458, + "upper": 0.0458, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.77, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5075, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.69, + "uncertainty": { + "confidence_interval": { + "lower": -0.0453, + "upper": 0.0453, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/deepseek/deepseek-v3.1/82ae1a6a-0f6b-4cd2-8196-dc422a8aeaac.json b/data/global-mmlu-lite/deepseek/deepseek-v3.1/82ae1a6a-0f6b-4cd2-8196-dc422a8aeaac.json new file mode 100644 index 0000000000000000000000000000000000000000..cd9e6cf27592f7230d46026c225a4148268a04dd --- /dev/null +++ b/data/global-mmlu-lite/deepseek/deepseek-v3.1/82ae1a6a-0f6b-4cd2-8196-dc422a8aeaac.json @@ -0,0 +1,519 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/deepseek_deepseek-v3.1/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "deepseek-v3.1", + "id": "deepseek/deepseek-v3.1", + "developer": "deepseek", + "inference_platform": "unknown" + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8044 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7793 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8295 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.805, + "uncertainty": { + "confidence_interval": { + "lower": -0.0388, + "upper": 0.0388, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0372, + "upper": 0.0372, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8157, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8175, + "uncertainty": { + "confidence_interval": { + "lower": -0.0379, + "upper": 0.0379, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7569, + "uncertainty": { + "confidence_interval": { + "lower": -0.0421, + "upper": 0.0421, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7764, + "uncertainty": { + "confidence_interval": { + "lower": -0.0409, + "upper": 0.0409, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0386, + "upper": 0.0386, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8312, + "uncertainty": { + "confidence_interval": { + "lower": -0.0374, + "upper": 0.0374, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8246, + "uncertainty": { + "confidence_interval": { + "lower": -0.0373, + "upper": 0.0373, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.801, + "uncertainty": { + "confidence_interval": { + "lower": -0.0393, + "upper": 0.0393, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7831, + "uncertainty": { + "confidence_interval": { + "lower": -0.0415, + "upper": 0.0415, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8161, + "uncertainty": { + "confidence_interval": { + "lower": -0.0381, + "upper": 0.0381, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemini-2.5-flash-preview-05-20/3355a584-4bcb-449d-ad27-27ab51bb37d8.json b/data/global-mmlu-lite/google/gemini-2.5-flash-preview-05-20/3355a584-4bcb-449d-ad27-27ab51bb37d8.json new file mode 100644 index 0000000000000000000000000000000000000000..7a752b15f7cfeb68bb9cc6eabdd404888a5932da --- /dev/null +++ b/data/global-mmlu-lite/google/gemini-2.5-flash-preview-05-20/3355a584-4bcb-449d-ad27-27ab51bb37d8.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemini-2.5-flash-preview-05-20/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemini-2.5-flash-preview-05-20", + "id": "google/gemini-2.5-flash-preview-05-20", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemini 2.5 Flash Preview" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9092 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8925 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9259 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9225, + "uncertainty": { + "confidence_interval": { + "lower": -0.0262, + "upper": 0.0262, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0258, + "upper": 0.0258, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.89, + "uncertainty": { + "confidence_interval": { + "lower": -0.0307, + "upper": 0.0307, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0316, + "upper": 0.0316, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.93, + "uncertainty": { + "confidence_interval": { + "lower": -0.025, + "upper": 0.025, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemini-2.5-flash/92f2b2f7-24b1-45cc-b607-0f6b68a5958f.json b/data/global-mmlu-lite/google/gemini-2.5-flash/92f2b2f7-24b1-45cc-b607-0f6b68a5958f.json new file mode 100644 index 0000000000000000000000000000000000000000..63fd35c63680acb8ff3f00377aa5b934d3885ab0 --- /dev/null +++ b/data/global-mmlu-lite/google/gemini-2.5-flash/92f2b2f7-24b1-45cc-b607-0f6b68a5958f.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemini-2.5-flash/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemini-2.5-flash", + "id": "google/gemini-2.5-flash", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemini 2.5 Flash" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9145 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9291 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9325, + "uncertainty": { + "confidence_interval": { + "lower": -0.0246, + "upper": 0.0246, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0258, + "upper": 0.0258, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9225, + "uncertainty": { + "confidence_interval": { + "lower": -0.0262, + "upper": 0.0262, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9175, + "uncertainty": { + "confidence_interval": { + "lower": -0.027, + "upper": 0.027, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemini-2.5-pro/cf06df61-6d21-4f86-83a1-41eeae596e48.json b/data/global-mmlu-lite/google/gemini-2.5-pro/cf06df61-6d21-4f86-83a1-41eeae596e48.json new file mode 100644 index 0000000000000000000000000000000000000000..9b089c6047af6b4cd06d9519f1525941a5497b34 --- /dev/null +++ b/data/global-mmlu-lite/google/gemini-2.5-pro/cf06df61-6d21-4f86-83a1-41eeae596e48.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemini-2.5-pro/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemini-2.5-pro", + "id": "google/gemini-2.5-pro", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemini 2.5 Pro" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9323 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9241 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9406 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0254, + "upper": 0.0254, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0254, + "upper": 0.0254, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.93, + "uncertainty": { + "confidence_interval": { + "lower": -0.025, + "upper": 0.025, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0254, + "upper": 0.0254, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0258, + "upper": 0.0258, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.935, + "uncertainty": { + "confidence_interval": { + "lower": -0.0242, + "upper": 0.0242, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0237, + "upper": 0.0237, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0254, + "upper": 0.0254, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.93, + "uncertainty": { + "confidence_interval": { + "lower": -0.025, + "upper": 0.025, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0237, + "upper": 0.0237, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0258, + "upper": 0.0258, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0254, + "upper": 0.0254, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.93, + "uncertainty": { + "confidence_interval": { + "lower": -0.025, + "upper": 0.025, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemini-3-pro-preview/f4639a49-e254-4dc2-906b-d9466457e7c6.json b/data/global-mmlu-lite/google/gemini-3-pro-preview/f4639a49-e254-4dc2-906b-d9466457e7c6.json new file mode 100644 index 0000000000000000000000000000000000000000..75c6588c0757f7272b3021eea375c136a6019118 --- /dev/null +++ b/data/global-mmlu-lite/google/gemini-3-pro-preview/f4639a49-e254-4dc2-906b-d9466457e7c6.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemini-3-pro-preview/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemini-3-pro-preview", + "id": "google/gemini-3-pro-preview", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemini 3 Pro Preview" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9453 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9397 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9509 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0198, + "upper": 0.0198, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.955, + "uncertainty": { + "confidence_interval": { + "lower": -0.0203, + "upper": 0.0203, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.955, + "uncertainty": { + "confidence_interval": { + "lower": -0.0203, + "upper": 0.0203, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94, + "uncertainty": { + "confidence_interval": { + "lower": -0.0233, + "upper": 0.0233, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemini-3.5-flash/8a9501d2-68fa-40d0-8684-4a66ab5c5da8.json b/data/global-mmlu-lite/google/gemini-3.5-flash/8a9501d2-68fa-40d0-8684-4a66ab5c5da8.json new file mode 100644 index 0000000000000000000000000000000000000000..d43a4dccbea6bb78470393731af2961108cc7b66 --- /dev/null +++ b/data/global-mmlu-lite/google/gemini-3.5-flash/8a9501d2-68fa-40d0-8684-4a66ab5c5da8.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemini-3.5-flash/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemini-3.5-flash", + "id": "google/gemini-3.5-flash", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemini 3.5 Flash" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9539 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9447 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9631 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.96, + "uncertainty": { + "confidence_interval": { + "lower": -0.0192, + "upper": 0.0192, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.96, + "uncertainty": { + "confidence_interval": { + "lower": -0.0192, + "upper": 0.0192, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0219, + "upper": 0.0219, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0198, + "upper": 0.0198, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0198, + "upper": 0.0198, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0228, + "upper": 0.0228, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9599, + "uncertainty": { + "confidence_interval": { + "lower": -0.0193, + "upper": 0.0193, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0198, + "upper": 0.0198, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0208, + "upper": 0.0208, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.955, + "uncertainty": { + "confidence_interval": { + "lower": -0.0203, + "upper": 0.0203, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.95, + "uncertainty": { + "confidence_interval": { + "lower": -0.0214, + "upper": 0.0214, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945, + "uncertainty": { + "confidence_interval": { + "lower": -0.0223, + "upper": 0.0223, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.96, + "uncertainty": { + "confidence_interval": { + "lower": -0.0192, + "upper": 0.0192, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0208, + "upper": 0.0208, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.955, + "uncertainty": { + "confidence_interval": { + "lower": -0.0203, + "upper": 0.0203, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemma-3-27b-it/827e6301-a7cd-44af-b981-41d13580fc4b.json b/data/global-mmlu-lite/google/gemma-3-27b-it/827e6301-a7cd-44af-b981-41d13580fc4b.json new file mode 100644 index 0000000000000000000000000000000000000000..7b5cc38073eadb8882dc2f9be5d14fbd2a26a616 --- /dev/null +++ b/data/global-mmlu-lite/google/gemma-3-27b-it/827e6301-a7cd-44af-b981-41d13580fc4b.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemma-3-27b-it/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemma-3-27b-it", + "id": "google/gemma-3-27b-it", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemma 3 27B" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.763 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7528 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7733 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.78, + "uncertainty": { + "confidence_interval": { + "lower": -0.0406, + "upper": 0.0406, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7337, + "uncertainty": { + "confidence_interval": { + "lower": -0.0434, + "upper": 0.0434, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.75, + "uncertainty": { + "confidence_interval": { + "lower": -0.0426, + "upper": 0.0426, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0409, + "upper": 0.0409, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7481, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7335, + "uncertainty": { + "confidence_interval": { + "lower": -0.0437, + "upper": 0.0437, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7563, + "uncertainty": { + "confidence_interval": { + "lower": -0.0422, + "upper": 0.0422, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.75, + "uncertainty": { + "confidence_interval": { + "lower": -0.0424, + "upper": 0.0424, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.798, + "uncertainty": { + "confidence_interval": { + "lower": -0.0395, + "upper": 0.0395, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7481, + "uncertainty": { + "confidence_interval": { + "lower": -0.0427, + "upper": 0.0427, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7494, + "uncertainty": { + "confidence_interval": { + "lower": -0.0425, + "upper": 0.0425, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.785, + "uncertainty": { + "confidence_interval": { + "lower": -0.0403, + "upper": 0.0403, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7444, + "uncertainty": { + "confidence_interval": { + "lower": -0.0428, + "upper": 0.0428, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7719, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemma-3-4b-it/3d4a62a8-a402-49a2-bb72-88d7d0eb342b.json b/data/global-mmlu-lite/google/gemma-3-4b-it/3d4a62a8-a402-49a2-bb72-88d7d0eb342b.json new file mode 100644 index 0000000000000000000000000000000000000000..a97c93918cd99f8b329d04f0e54b34b83bfa1852 --- /dev/null +++ b/data/global-mmlu-lite/google/gemma-3-4b-it/3d4a62a8-a402-49a2-bb72-88d7d0eb342b.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemma-3-4b-it/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemma-3-4b-it", + "id": "google/gemma-3-4b-it", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemma 3 4B" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6511 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6116 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6906 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0467, + "upper": 0.0467, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.67, + "uncertainty": { + "confidence_interval": { + "lower": -0.0461, + "upper": 0.0461, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.68, + "uncertainty": { + "confidence_interval": { + "lower": -0.0457, + "upper": 0.0457, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0467, + "upper": 0.0467, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0468, + "upper": 0.0468, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0458, + "upper": 0.0458, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0462, + "upper": 0.0462, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6325, + "uncertainty": { + "confidence_interval": { + "lower": -0.0472, + "upper": 0.0472, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.66, + "uncertainty": { + "confidence_interval": { + "lower": -0.0464, + "upper": 0.0464, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.68, + "uncertainty": { + "confidence_interval": { + "lower": -0.0457, + "upper": 0.0457, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6725, + "uncertainty": { + "confidence_interval": { + "lower": -0.046, + "upper": 0.046, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0479, + "upper": 0.0479, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0483, + "upper": 0.0483, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0468, + "upper": 0.0468, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.63, + "uncertainty": { + "confidence_interval": { + "lower": -0.0473, + "upper": 0.0473, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemma-4-26b-a4b-it/ab04e613-6e9c-4682-9723-132b46ed0738.json b/data/global-mmlu-lite/google/gemma-4-26b-a4b-it/ab04e613-6e9c-4682-9723-132b46ed0738.json new file mode 100644 index 0000000000000000000000000000000000000000..2eb6546b524a8d74ec809bd001bff94aca24f3c9 --- /dev/null +++ b/data/global-mmlu-lite/google/gemma-4-26b-a4b-it/ab04e613-6e9c-4682-9723-132b46ed0738.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemma-4-26b-a4b-it/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemma-4-26b-a4b-it", + "id": "google/gemma-4-26b-a4b-it", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemma 4 26B A4B" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8571 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8465 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8677 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8596, + "uncertainty": { + "confidence_interval": { + "lower": -0.0341, + "upper": 0.0341, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8797, + "uncertainty": { + "confidence_interval": { + "lower": -0.0319, + "upper": 0.0319, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8308, + "uncertainty": { + "confidence_interval": { + "lower": -0.0369, + "upper": 0.0369, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8596, + "uncertainty": { + "confidence_interval": { + "lower": -0.0341, + "upper": 0.0341, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8546, + "uncertainty": { + "confidence_interval": { + "lower": -0.0346, + "upper": 0.0346, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0357, + "upper": 0.0357, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8521, + "uncertainty": { + "confidence_interval": { + "lower": -0.0348, + "upper": 0.0348, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8546, + "uncertainty": { + "confidence_interval": { + "lower": -0.0346, + "upper": 0.0346, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8722, + "uncertainty": { + "confidence_interval": { + "lower": -0.0328, + "upper": 0.0328, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8766, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8467, + "uncertainty": { + "confidence_interval": { + "lower": -0.0354, + "upper": 0.0354, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0332, + "upper": 0.0332, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.855, + "uncertainty": { + "confidence_interval": { + "lower": -0.0345, + "upper": 0.0345, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8246, + "uncertainty": { + "confidence_interval": { + "lower": -0.0373, + "upper": 0.0373, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8797, + "uncertainty": { + "confidence_interval": { + "lower": -0.0319, + "upper": 0.0319, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8571, + "uncertainty": { + "confidence_interval": { + "lower": -0.0343, + "upper": 0.0343, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/google/gemma-4-31b-it/2dc4844b-2ada-42a7-9d46-197b50359bc7.json b/data/global-mmlu-lite/google/gemma-4-31b-it/2dc4844b-2ada-42a7-9d46-197b50359bc7.json new file mode 100644 index 0000000000000000000000000000000000000000..0c52956e96bf816078817b6725340e8451be942b --- /dev/null +++ b/data/global-mmlu-lite/google/gemma-4-31b-it/2dc4844b-2ada-42a7-9d46-197b50359bc7.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/google_gemma-4-31b-it/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gemma-4-31b-it", + "id": "google/gemma-4-31b-it", + "developer": "google", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Gemma 4 31B" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9068 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8925 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9212 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9175, + "uncertainty": { + "confidence_interval": { + "lower": -0.027, + "upper": 0.027, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9, + "uncertainty": { + "confidence_interval": { + "lower": -0.0294, + "upper": 0.0294, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.92, + "uncertainty": { + "confidence_interval": { + "lower": -0.0266, + "upper": 0.0266, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.902, + "uncertainty": { + "confidence_interval": { + "lower": -0.0292, + "upper": 0.0292, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0297, + "upper": 0.0297, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0297, + "upper": 0.0297, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/ibm/granite-4.0-h-small/ec935f95-9930-4f99-ac8d-03de3c244dda.json b/data/global-mmlu-lite/ibm/granite-4.0-h-small/ec935f95-9930-4f99-ac8d-03de3c244dda.json new file mode 100644 index 0000000000000000000000000000000000000000..e7cf014269d4a4dc37587685c8cad97b6c6300f0 --- /dev/null +++ b/data/global-mmlu-lite/ibm/granite-4.0-h-small/ec935f95-9930-4f99-ac8d-03de3c244dda.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/ibm_granite-4.0-h-small/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "granite-4.0-h-small", + "id": "ibm/granite-4.0-h-small", + "developer": "ibm", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Granite 4.0 Small" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7503 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7182 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7826 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7613, + "uncertainty": { + "confidence_interval": { + "lower": -0.0419, + "upper": 0.0419, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.77, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7613, + "uncertainty": { + "confidence_interval": { + "lower": -0.0419, + "upper": 0.0419, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.755, + "uncertainty": { + "confidence_interval": { + "lower": -0.0421, + "upper": 0.0421, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7594, + "uncertainty": { + "confidence_interval": { + "lower": -0.0419, + "upper": 0.0419, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7575, + "uncertainty": { + "confidence_interval": { + "lower": -0.042, + "upper": 0.042, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7614, + "uncertainty": { + "confidence_interval": { + "lower": -0.0421, + "upper": 0.0421, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7406, + "uncertainty": { + "confidence_interval": { + "lower": -0.0431, + "upper": 0.0431, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.757, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7638, + "uncertainty": { + "confidence_interval": { + "lower": -0.0417, + "upper": 0.0417, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7318, + "uncertainty": { + "confidence_interval": { + "lower": -0.0435, + "upper": 0.0435, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6921, + "uncertainty": { + "confidence_interval": { + "lower": -0.0456, + "upper": 0.0456, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7475, + "uncertainty": { + "confidence_interval": { + "lower": -0.0426, + "upper": 0.0426, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7419, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/mistralai/mistral-medium-3/6a672cfa-d357-45cb-9ab9-16d5346fcab0.json b/data/global-mmlu-lite/mistralai/mistral-medium-3/6a672cfa-d357-45cb-9ab9-16d5346fcab0.json new file mode 100644 index 0000000000000000000000000000000000000000..a8a1378f7e1b97415bbd509e02952d5a02a32480 --- /dev/null +++ b/data/global-mmlu-lite/mistralai/mistral-medium-3/6a672cfa-d357-45cb-9ab9-16d5346fcab0.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/mistralai_mistral-medium-3/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "mistral-medium-3", + "id": "mistralai/mistral-medium-3", + "developer": "mistralai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Mistral Medium 3" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5511 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5391 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5631 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.455, + "uncertainty": { + "confidence_interval": { + "lower": -0.0488, + "upper": 0.0488, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.38, + "uncertainty": { + "confidence_interval": { + "lower": -0.0476, + "upper": 0.0476, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5175, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0489, + "upper": 0.0489, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.41, + "uncertainty": { + "confidence_interval": { + "lower": -0.0482, + "upper": 0.0482, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.555, + "uncertainty": { + "confidence_interval": { + "lower": -0.0487, + "upper": 0.0487, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.515, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.535, + "uncertainty": { + "confidence_interval": { + "lower": -0.0489, + "upper": 0.0489, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.58, + "uncertainty": { + "confidence_interval": { + "lower": -0.0484, + "upper": 0.0484, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.595, + "uncertainty": { + "confidence_interval": { + "lower": -0.0481, + "upper": 0.0481, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5175, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5375, + "uncertainty": { + "confidence_interval": { + "lower": -0.0489, + "upper": 0.0489, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0446, + "upper": 0.0446, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0414, + "upper": 0.0414, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.535, + "uncertainty": { + "confidence_interval": { + "lower": -0.0489, + "upper": 0.0489, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7325, + "uncertainty": { + "confidence_interval": { + "lower": -0.0434, + "upper": 0.0434, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/mistralai/mistral-small-2503/653eafbb-5405-4b59-a20b-0adcdfa0c561.json b/data/global-mmlu-lite/mistralai/mistral-small-2503/653eafbb-5405-4b59-a20b-0adcdfa0c561.json new file mode 100644 index 0000000000000000000000000000000000000000..e75bfe56aebc5b684438508c5dc49f494de6456e --- /dev/null +++ b/data/global-mmlu-lite/mistralai/mistral-small-2503/653eafbb-5405-4b59-a20b-0adcdfa0c561.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/mistralai_mistral-small-2503/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "mistral-small-2503", + "id": "mistralai/mistral-small-2503", + "developer": "mistralai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Mistral Small 3.1" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7852 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7537 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8166 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0401, + "upper": 0.0401, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8, + "uncertainty": { + "confidence_interval": { + "lower": -0.0392, + "upper": 0.0392, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0411, + "upper": 0.0411, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0394, + "upper": 0.0394, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8, + "uncertainty": { + "confidence_interval": { + "lower": -0.0392, + "upper": 0.0392, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.795, + "uncertainty": { + "confidence_interval": { + "lower": -0.0396, + "upper": 0.0396, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.785, + "uncertainty": { + "confidence_interval": { + "lower": -0.0403, + "upper": 0.0403, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.805, + "uncertainty": { + "confidence_interval": { + "lower": -0.0388, + "upper": 0.0388, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.77, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.79, + "uncertainty": { + "confidence_interval": { + "lower": -0.0399, + "upper": 0.0399, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0404, + "upper": 0.0404, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0409, + "upper": 0.0409, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.735, + "uncertainty": { + "confidence_interval": { + "lower": -0.0432, + "upper": 0.0432, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0397, + "upper": 0.0397, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0404, + "upper": 0.0404, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/openai/gpt-4.1-2025-04-14/283f51d5-7745-438b-9f70-792740176eba.json b/data/global-mmlu-lite/openai/gpt-4.1-2025-04-14/283f51d5-7745-438b-9f70-792740176eba.json new file mode 100644 index 0000000000000000000000000000000000000000..c0884d38205db17e766cbc76e6cf7ac7fa4247aa --- /dev/null +++ b/data/global-mmlu-lite/openai/gpt-4.1-2025-04-14/283f51d5-7745-438b-9f70-792740176eba.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/openai_gpt-4.1-2025-04-14/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gpt-4.1-2025-04-14", + "id": "openai/gpt-4.1-2025-04-14", + "developer": "openai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "GPT-4.1" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8755 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8541 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8969 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8825, + "uncertainty": { + "confidence_interval": { + "lower": -0.0316, + "upper": 0.0316, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8625, + "uncertainty": { + "confidence_interval": { + "lower": -0.0337, + "upper": 0.0337, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0321, + "upper": 0.0321, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885, + "uncertainty": { + "confidence_interval": { + "lower": -0.0313, + "upper": 0.0313, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885, + "uncertainty": { + "confidence_interval": { + "lower": -0.0313, + "upper": 0.0313, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0343, + "upper": 0.0343, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/openai/gpt-5-2025-08-07/51afecd5-44e4-4056-bdc2-29b29d8a3cb5.json b/data/global-mmlu-lite/openai/gpt-5-2025-08-07/51afecd5-44e4-4056-bdc2-29b29d8a3cb5.json new file mode 100644 index 0000000000000000000000000000000000000000..3e5a4e3c4d8af3d9130c3fbbac7d109973bbb1ce --- /dev/null +++ b/data/global-mmlu-lite/openai/gpt-5-2025-08-07/51afecd5-44e4-4056-bdc2-29b29d8a3cb5.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/openai_gpt-5-2025-08-07/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "gpt-5-2025-08-07", + "id": "openai/gpt-5-2025-08-07", + "developer": "openai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "GPT-5" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8895 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8913 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8878 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0304, + "upper": 0.0304, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9, + "uncertainty": { + "confidence_interval": { + "lower": -0.0294, + "upper": 0.0294, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.865, + "uncertainty": { + "confidence_interval": { + "lower": -0.0335, + "upper": 0.0335, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.795, + "uncertainty": { + "confidence_interval": { + "lower": -0.0396, + "upper": 0.0396, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.865, + "uncertainty": { + "confidence_interval": { + "lower": -0.0335, + "upper": 0.0335, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0277, + "upper": 0.0277, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.895, + "uncertainty": { + "confidence_interval": { + "lower": -0.03, + "upper": 0.03, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.915, + "uncertainty": { + "confidence_interval": { + "lower": -0.0273, + "upper": 0.0273, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/openai/o3-mini-2025-01-31/68ebbee4-6e28-467f-aca7-12dda3321470.json b/data/global-mmlu-lite/openai/o3-mini-2025-01-31/68ebbee4-6e28-467f-aca7-12dda3321470.json new file mode 100644 index 0000000000000000000000000000000000000000..8a78fcee42ea98877e5cc9634e15365d3bc4357a --- /dev/null +++ b/data/global-mmlu-lite/openai/o3-mini-2025-01-31/68ebbee4-6e28-467f-aca7-12dda3321470.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/openai_o3-mini-2025-01-31/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "o3-mini-2025-01-31", + "id": "openai/o3-mini-2025-01-31", + "developer": "openai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "o3 mini" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.78 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.765 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.795 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0411, + "upper": 0.0411, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8025, + "uncertainty": { + "confidence_interval": { + "lower": -0.039, + "upper": 0.039, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.77, + "uncertainty": { + "confidence_interval": { + "lower": -0.0412, + "upper": 0.0412, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.74, + "uncertainty": { + "confidence_interval": { + "lower": -0.043, + "upper": 0.043, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0423, + "upper": 0.0423, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7425, + "uncertainty": { + "confidence_interval": { + "lower": -0.0429, + "upper": 0.0429, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8, + "uncertainty": { + "confidence_interval": { + "lower": -0.0392, + "upper": 0.0392, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.81, + "uncertainty": { + "confidence_interval": { + "lower": -0.0384, + "upper": 0.0384, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0386, + "upper": 0.0386, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7975, + "uncertainty": { + "confidence_interval": { + "lower": -0.0394, + "upper": 0.0394, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0409, + "upper": 0.0409, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.765, + "uncertainty": { + "confidence_interval": { + "lower": -0.0416, + "upper": 0.0416, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0411, + "upper": 0.0411, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8125, + "uncertainty": { + "confidence_interval": { + "lower": -0.0382, + "upper": 0.0382, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0386, + "upper": 0.0386, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/openai/o4-mini-2025-04-16/350a7b7c-fb17-42b6-8c16-1c220f2863c0.json b/data/global-mmlu-lite/openai/o4-mini-2025-04-16/350a7b7c-fb17-42b6-8c16-1c220f2863c0.json new file mode 100644 index 0000000000000000000000000000000000000000..1d9d5ae0e3309168c75f852836debadd81619fe2 --- /dev/null +++ b/data/global-mmlu-lite/openai/o4-mini-2025-04-16/350a7b7c-fb17-42b6-8c16-1c220f2863c0.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/openai_o4-mini-2025-04-16/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "o4-mini-2025-04-16", + "id": "openai/o4-mini-2025-04-16", + "developer": "openai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "o4 mini" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8705 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8503 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8906 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.865, + "uncertainty": { + "confidence_interval": { + "lower": -0.0335, + "upper": 0.0335, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0332, + "upper": 0.0332, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8875, + "uncertainty": { + "confidence_interval": { + "lower": -0.031, + "upper": 0.031, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8775, + "uncertainty": { + "confidence_interval": { + "lower": -0.0321, + "upper": 0.0321, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0332, + "upper": 0.0332, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.855, + "uncertainty": { + "confidence_interval": { + "lower": -0.0345, + "upper": 0.0345, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885, + "uncertainty": { + "confidence_interval": { + "lower": -0.0313, + "upper": 0.0313, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88, + "uncertainty": { + "confidence_interval": { + "lower": -0.0318, + "upper": 0.0318, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.855, + "uncertainty": { + "confidence_interval": { + "lower": -0.0345, + "upper": 0.0345, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0348, + "upper": 0.0348, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0348, + "upper": 0.0348, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.89, + "uncertainty": { + "confidence_interval": { + "lower": -0.0307, + "upper": 0.0307, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/xai/grok-3-mini/a8bb2b5c-5f6f-42db-8c66-f9eb894f99c0.json b/data/global-mmlu-lite/xai/grok-3-mini/a8bb2b5c-5f6f-42db-8c66-f9eb894f99c0.json new file mode 100644 index 0000000000000000000000000000000000000000..089bd12d102b61ed4e92ab90dab952fd89f44761 --- /dev/null +++ b/data/global-mmlu-lite/xai/grok-3-mini/a8bb2b5c-5f6f-42db-8c66-f9eb894f99c0.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/xai_grok-3-mini/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "grok-3-mini", + "id": "xai/grok-3-mini", + "developer": "xai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Grok 3 Mini" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.673 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6717 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6743 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.755, + "uncertainty": { + "confidence_interval": { + "lower": -0.0421, + "upper": 0.0421, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5075, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7355, + "uncertainty": { + "confidence_interval": { + "lower": -0.0434, + "upper": 0.0434, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6591, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.485, + "uncertainty": { + "confidence_interval": { + "lower": -0.049, + "upper": 0.049, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.56, + "uncertainty": { + "confidence_interval": { + "lower": -0.0486, + "upper": 0.0486, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0438, + "upper": 0.0438, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.696, + "uncertainty": { + "confidence_interval": { + "lower": -0.0452, + "upper": 0.0452, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6575, + "uncertainty": { + "confidence_interval": { + "lower": -0.0465, + "upper": 0.0465, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7325, + "uncertainty": { + "confidence_interval": { + "lower": -0.0434, + "upper": 0.0434, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6275, + "uncertainty": { + "confidence_interval": { + "lower": -0.0474, + "upper": 0.0474, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.61, + "uncertainty": { + "confidence_interval": { + "lower": -0.0478, + "upper": 0.0478, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7625, + "uncertainty": { + "confidence_interval": { + "lower": -0.0417, + "upper": 0.0417, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8296, + "uncertainty": { + "confidence_interval": { + "lower": -0.0369, + "upper": 0.0369, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5564, + "uncertainty": { + "confidence_interval": { + "lower": -0.0487, + "upper": 0.0487, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8693, + "uncertainty": { + "confidence_interval": { + "lower": -0.0331, + "upper": 0.0331, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/global-mmlu-lite/xai/grok-4-0709/b6665653-b52f-43d5-9d63-c7a574d1ff9c.json b/data/global-mmlu-lite/xai/grok-4-0709/b6665653-b52f-43d5-9d63-c7a574d1ff9c.json new file mode 100644 index 0000000000000000000000000000000000000000..4d1795d727e457ec33c78f342d7067e17ef0e4c9 --- /dev/null +++ b/data/global-mmlu-lite/xai/grok-4-0709/b6665653-b52f-43d5-9d63-c7a574d1ff9c.json @@ -0,0 +1,522 @@ +{ + "schema_version": "0.2.2", + "evaluation_id": "global-mmlu-lite/xai_grok-4-0709/1783820827.6963763", + "retrieved_timestamp": "1783820827.6963763", + "source_metadata": { + "source_name": "Global MMLU Lite Leaderboard", + "source_type": "documentation", + "source_organization_name": "kaggle", + "source_organization_url": "www.kaggle.com", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "kaggle kernel", + "version": "4", + "additional_details": { + "url": "https://www.kaggle.com/code/shivalikasingh95/global-mmlu-lite-sample-notebook" + } + }, + "model_info": { + "name": "grok-4-0709", + "id": "xai/grok-4-0709", + "developer": "xai", + "inference_platform": "unknown", + "additional_details": { + "display_name": "Grok 4" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Average", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Average", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8881 + } + }, + { + "evaluation_name": "Culturally Sensitive", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Sensitive", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8862 + } + }, + { + "evaluation_name": "Culturally Agnostic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Culturally Agnostic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.89 + } + }, + { + "evaluation_name": "Arabic", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Arabic", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885, + "uncertainty": { + "confidence_interval": { + "lower": -0.0313, + "upper": 0.0313, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "English", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - English", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Bengali", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Bengali", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8925, + "uncertainty": { + "confidence_interval": { + "lower": -0.0304, + "upper": 0.0304, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "German", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - German", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "French", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - French", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875, + "uncertainty": { + "confidence_interval": { + "lower": -0.0324, + "upper": 0.0324, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Hindi", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Hindi", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8675, + "uncertainty": { + "confidence_interval": { + "lower": -0.0332, + "upper": 0.0332, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Indonesian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Indonesian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.89, + "uncertainty": { + "confidence_interval": { + "lower": -0.0307, + "upper": 0.0307, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Italian", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Italian", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9025, + "uncertainty": { + "confidence_interval": { + "lower": -0.0291, + "upper": 0.0291, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Japanese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Japanese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.87, + "uncertainty": { + "confidence_interval": { + "lower": -0.033, + "upper": 0.033, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Korean", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Korean", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.895, + "uncertainty": { + "confidence_interval": { + "lower": -0.03, + "upper": 0.03, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Portuguese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Portuguese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8725, + "uncertainty": { + "confidence_interval": { + "lower": -0.0327, + "upper": 0.0327, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Spanish", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Spanish", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Swahili", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Swahili", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.91, + "uncertainty": { + "confidence_interval": { + "lower": -0.028, + "upper": 0.028, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Yoruba", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Yoruba", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.905, + "uncertainty": { + "confidence_interval": { + "lower": -0.0287, + "upper": 0.0287, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Chinese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Chinese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8525, + "uncertainty": { + "confidence_interval": { + "lower": -0.0348, + "upper": 0.0348, + "method": "unknown" + } + } + } + }, + { + "evaluation_name": "Burmese", + "source_data": { + "dataset_name": "global-mmlu-lite", + "source_type": "url", + "url": [ + "https://www.kaggle.com/datasets/cohere-labs/global-mmlu-lite" + ] + }, + "metric_config": { + "evaluation_description": "Global MMLU Lite - Burmese", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9075, + "uncertainty": { + "confidence_interval": { + "lower": -0.0284, + "upper": 0.0284, + "method": "unknown" + } + } + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/0-hero/Matter-0.1-7B-DPO-preview/2fbdc84a-0777-4e21-9143-cb2d9c75cc20.json b/data/rewardbench/0-hero/Matter-0.1-7B-DPO-preview/2fbdc84a-0777-4e21-9143-cb2d9c75cc20.json new file mode 100644 index 0000000000000000000000000000000000000000..6e8a9b95d2295fb3b26155fc064e6496b4c51fab --- /dev/null +++ b/data/rewardbench/0-hero/Matter-0.1-7B-DPO-preview/2fbdc84a-0777-4e21-9143-cb2d9c75cc20.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/0-hero_Matter-0.1-7B-DPO-preview/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "0-hero/Matter-0.1-7B-DPO-preview", + "id": "0-hero/Matter-0.1-7B-DPO-preview", + "developer": "0-hero", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7247 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8939 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5768 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6378 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8854 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5348 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/0-hero/Matter-0.1-7B-boost-DPO-preview/ca0344bb-1c2b-4796-8b85-d326fc89a2aa.json b/data/rewardbench/0-hero/Matter-0.1-7B-boost-DPO-preview/ca0344bb-1c2b-4796-8b85-d326fc89a2aa.json new file mode 100644 index 0000000000000000000000000000000000000000..7a8aae23b9e00c6058287ee5246894af91a5c490 --- /dev/null +++ b/data/rewardbench/0-hero/Matter-0.1-7B-boost-DPO-preview/ca0344bb-1c2b-4796-8b85-d326fc89a2aa.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/0-hero_Matter-0.1-7B-boost-DPO-preview/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "0-hero/Matter-0.1-7B-boost-DPO-preview", + "id": "0-hero/Matter-0.1-7B-boost-DPO-preview", + "developer": "0-hero", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7448 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9106 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6096 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7135 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8395 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5566 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ahjeong/MMPO_Gemma_7b/66190e14-2862-4f53-8224-f968e4d55732.json b/data/rewardbench/Ahjeong/MMPO_Gemma_7b/66190e14-2862-4f53-8224-f968e4d55732.json new file mode 100644 index 0000000000000000000000000000000000000000..8f8fc951d6a36c1fc39e166a380d72c6e2f2ff75 --- /dev/null +++ b/data/rewardbench/Ahjeong/MMPO_Gemma_7b/66190e14-2862-4f53-8224-f968e4d55732.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ahjeong_MMPO_Gemma_7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ahjeong/MMPO_Gemma_7b", + "id": "Ahjeong/MMPO_Gemma_7b", + "developer": "Ahjeong", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7587 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.614 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7135 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7756 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6831 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3/e96f9bba-8efe-457e-a6b1-4f90b8739f17.json b/data/rewardbench/Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3/e96f9bba-8efe-457e-a6b1-4f90b8739f17.json new file mode 100644 index 0000000000000000000000000000000000000000..51921202f96a6abdcac7ff3a3fe71121368aa572 --- /dev/null +++ b/data/rewardbench/Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3/e96f9bba-8efe-457e-a6b1-4f90b8739f17.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ahjeong_MMPO_Gemma_7b_gamma1.1_epoch3/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3", + "id": "Ahjeong/MMPO_Gemma_7b_gamma1.1_epoch3", + "developer": "Ahjeong", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7652 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9721 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6338 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7635 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7284 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6913 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Anthropic/claude-3-5-sonnet-20240620/25fcb524-3e9b-4d97-b84d-01986b127c6c.json b/data/rewardbench/Anthropic/claude-3-5-sonnet-20240620/25fcb524-3e9b-4d97-b84d-01986b127c6c.json new file mode 100644 index 0000000000000000000000000000000000000000..2aae6a72a2fe84c39e4d30260ebb08714bdf9bb3 --- /dev/null +++ b/data/rewardbench/Anthropic/claude-3-5-sonnet-20240620/25fcb524-3e9b-4d97-b84d-01986b127c6c.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Anthropic_claude-3-5-sonnet-20240620/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Anthropic/claude-3-5-sonnet-20240620", + "id": "Anthropic/claude-3-5-sonnet-20240620", + "developer": "Anthropic", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8417 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7401 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8162 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8469 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Anthropic/claude-3-haiku-20240307/9421228b-a52b-4b36-a28b-2ad49f2dd5cb.json b/data/rewardbench/Anthropic/claude-3-haiku-20240307/9421228b-a52b-4b36-a28b-2ad49f2dd5cb.json new file mode 100644 index 0000000000000000000000000000000000000000..8c38b3f3074e282be5572d55f88a25976e985b0d --- /dev/null +++ b/data/rewardbench/Anthropic/claude-3-haiku-20240307/9421228b-a52b-4b36-a28b-2ad49f2dd5cb.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Anthropic_claude-3-haiku-20240307/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Anthropic/claude-3-haiku-20240307", + "id": "Anthropic/claude-3-haiku-20240307", + "developer": "Anthropic", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7289 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9274 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5197 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7953 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.706 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6635 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Anthropic/claude-3-opus-20240229/6595e39a-1c58-424f-acc5-c6bff657bc39.json b/data/rewardbench/Anthropic/claude-3-opus-20240229/6595e39a-1c58-424f-acc5-c6bff657bc39.json new file mode 100644 index 0000000000000000000000000000000000000000..d877ba32bad8349ad41fcb2b6865201d9b17f7bf --- /dev/null +++ b/data/rewardbench/Anthropic/claude-3-opus-20240229/6595e39a-1c58-424f-acc5-c6bff657bc39.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Anthropic_claude-3-opus-20240229/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Anthropic/claude-3-opus-20240229", + "id": "Anthropic/claude-3-opus-20240229", + "developer": "Anthropic", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8008 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9469 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6031 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8662 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7868 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Anthropic/claude-3-sonnet-20240229/473f6ca5-f833-4cec-8395-88172151dfc5.json b/data/rewardbench/Anthropic/claude-3-sonnet-20240229/473f6ca5-f833-4cec-8395-88172151dfc5.json new file mode 100644 index 0000000000000000000000000000000000000000..5f57c7729b69e518cf89f02f5e17db1f383ecfa2 --- /dev/null +++ b/data/rewardbench/Anthropic/claude-3-sonnet-20240229/473f6ca5-f833-4cec-8395-88172151dfc5.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Anthropic_claude-3-sonnet-20240229/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Anthropic/claude-3-sonnet-20240229", + "id": "Anthropic/claude-3-sonnet-20240229", + "developer": "Anthropic", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7458 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9344 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5658 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8169 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6907 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6963 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/AtlaAI/Selene-1-Mini-Llama-3.1-8B/d1f55fa2-64ba-481d-9bff-f82a6dd0ca45.json b/data/rewardbench/AtlaAI/Selene-1-Mini-Llama-3.1-8B/d1f55fa2-64ba-481d-9bff-f82a6dd0ca45.json new file mode 100644 index 0000000000000000000000000000000000000000..85eca7547262525d1486fba8e3cfbd5e6f745b94 --- /dev/null +++ b/data/rewardbench/AtlaAI/Selene-1-Mini-Llama-3.1-8B/d1f55fa2-64ba-481d-9bff-f82a6dd0ca45.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/AtlaAI_Selene-1-Mini-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "AtlaAI/Selene-1-Mini-Llama-3.1-8B", + "id": "AtlaAI/Selene-1-Mini-Llama-3.1-8B", + "developer": "AtlaAI", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8913 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9358 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7939 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8926 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9429 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/AtlaAI/Selene-1/fd470f89-4dc5-4a7c-aa6b-135d05a9aab6.json b/data/rewardbench/AtlaAI/Selene-1/fd470f89-4dc5-4a7c-aa6b-135d05a9aab6.json new file mode 100644 index 0000000000000000000000000000000000000000..a2847e203cab30cac4b18ed61d74fe9f8e391a9a --- /dev/null +++ b/data/rewardbench/AtlaAI/Selene-1/fd470f89-4dc5-4a7c-aa6b-135d05a9aab6.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/AtlaAI_Selene-1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "AtlaAI/Selene-1", + "id": "AtlaAI/Selene-1", + "developer": "AtlaAI", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9241 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9777 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8399 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9216 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9572 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/216e0174-bab4-40ee-9871-1be675d27445.json b/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/216e0174-bab4-40ee-9871-1be675d27445.json new file mode 100644 index 0000000000000000000000000000000000000000..259ad206d50ba3f9c3653f481985182076475102 --- /dev/null +++ b/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/216e0174-bab4-40ee-9871-1be675d27445.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/CIR-AMS_BTRM_Qwen2_7b_0613/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "CIR-AMS/BTRM_Qwen2_7b_0613", + "id": "CIR-AMS/BTRM_Qwen2_7b_0613", + "developer": "CIR-AMS", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5736 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5347 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5737 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6527 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/3c1e3213-e5ac-47b5-b54a-8d1cbab56e01.json b/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/3c1e3213-e5ac-47b5-b54a-8d1cbab56e01.json new file mode 100644 index 0000000000000000000000000000000000000000..e5f2bc8b7ba4c7757b4815a923f4cb0c7240de7a --- /dev/null +++ b/data/rewardbench/CIR-AMS/BTRM_Qwen2_7b_0613/3c1e3213-e5ac-47b5-b54a-8d1cbab56e01.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/CIR-AMS_BTRM_Qwen2_7b_0613/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "CIR-AMS/BTRM_Qwen2_7b_0613", + "id": "CIR-AMS/BTRM_Qwen2_7b_0613", + "developer": "CIR-AMS", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8172 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5724 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9014 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8775 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7029 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/CohereForAI/c4ai-command-r-plus/1468c630-4d3a-42b4-b215-c6a457e9d452.json b/data/rewardbench/CohereForAI/c4ai-command-r-plus/1468c630-4d3a-42b4-b215-c6a457e9d452.json new file mode 100644 index 0000000000000000000000000000000000000000..3a36dd607062ca2c840600c33a742f0efcd1befa --- /dev/null +++ b/data/rewardbench/CohereForAI/c4ai-command-r-plus/1468c630-4d3a-42b4-b215-c6a457e9d452.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/CohereForAI_c4ai-command-r-plus/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "CohereForAI/c4ai-command-r-plus", + "id": "CohereForAI/c4ai-command-r-plus", + "developer": "CohereForAI", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7057 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9511 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5757 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5986 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.704 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6924 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/LMUnit-llama3.1-70b/712e97f2-b937-44fd-a27e-6f080a3091da.json b/data/rewardbench/ContextualAI/LMUnit-llama3.1-70b/712e97f2-b937-44fd-a27e-6f080a3091da.json new file mode 100644 index 0000000000000000000000000000000000000000..0c23e0eca44392d0ff6e3de50fa5cbc51dbe6d5d --- /dev/null +++ b/data/rewardbench/ContextualAI/LMUnit-llama3.1-70b/712e97f2-b937-44fd-a27e-6f080a3091da.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/ContextualAI_LMUnit-llama3.1-70b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/LMUnit-llama3.1-70b", + "id": "ContextualAI/LMUnit-llama3.1-70b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8054 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8463 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9067 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9697 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9063 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/LMUnit-qwen2.5-72b/9c657c91-2209-44ca-8bc0-48ec8d7e9bba.json b/data/rewardbench/ContextualAI/LMUnit-qwen2.5-72b/9c657c91-2209-44ca-8bc0-48ec8d7e9bba.json new file mode 100644 index 0000000000000000000000000000000000000000..a8b9964be3257c926b7bfea51510d4782d0e937d --- /dev/null +++ b/data/rewardbench/ContextualAI/LMUnit-qwen2.5-72b/9c657c91-2209-44ca-8bc0-48ec8d7e9bba.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/ContextualAI_LMUnit-qwen2.5-72b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/LMUnit-qwen2.5-72b", + "id": "ContextualAI/LMUnit-qwen2.5-72b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8208 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8716 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7268 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9133 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9677 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9014 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_llama13b/19e04aeb-d489-48b2-a5ae-2c837f23f686.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama13b/19e04aeb-d489-48b2-a5ae-2c837f23f686.json new file mode 100644 index 0000000000000000000000000000000000000000..47be7e5ae0b7cf217447499db3453b5432cdd32d --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama13b/19e04aeb-d489-48b2-a5ae-2c837f23f686.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama13b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_llama13b", + "id": "ContextualAI/archangel_sft-dpo_llama13b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.54 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7123 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4298 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5649 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4401 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5656 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_llama30b/4e7dbb9b-bd79-4e48-8b9a-4a19312e526a.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama30b/4e7dbb9b-bd79-4e48-8b9a-4a19312e526a.json new file mode 100644 index 0000000000000000000000000000000000000000..2689c3686eed1c109d941610e42c559984bd1a1a --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama30b/4e7dbb9b-bd79-4e48-8b9a-4a19312e526a.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama30b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_llama30b", + "id": "ContextualAI/archangel_sft-dpo_llama30b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5618 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6927 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4474 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4745 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5705 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_llama7b/2ffd133b-b4c2-4005-8166-42d1012b8397.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama7b/2ffd133b-b4c2-4005-8166-42d1012b8397.json new file mode 100644 index 0000000000000000000000000000000000000000..fc5bc5a944d84197426b80c4e5d01b5555549f79 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_llama7b/2ffd133b-b4c2-4005-8166-42d1012b8397.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_llama7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_llama7b", + "id": "ContextualAI/archangel_sft-dpo_llama7b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5304 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5782 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4452 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5203 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5658 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5544 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia1-4b/4a4b751a-ec7a-4d99-a0c2-f008cfc389a1.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia1-4b/4a4b751a-ec7a-4d99-a0c2-f008cfc389a1.json new file mode 100644 index 0000000000000000000000000000000000000000..d7cb2d40f8accf47b4e94f8a312feaa7466ea537 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia1-4b/4a4b751a-ec7a-4d99-a0c2-f008cfc389a1.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia1-4b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_pythia1-4b", + "id": "ContextualAI/archangel_sft-dpo_pythia1-4b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5233 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6397 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3728 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5041 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5672 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5427 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia12-0b/48a9c2c0-24f9-467d-9e2d-aabc0a8abdd1.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia12-0b/48a9c2c0-24f9-467d-9e2d-aabc0a8abdd1.json new file mode 100644 index 0000000000000000000000000000000000000000..1b6467c55cb084e021fca92f8e751dedf1124553 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia12-0b/48a9c2c0-24f9-467d-9e2d-aabc0a8abdd1.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia12-0b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_pythia12-0b", + "id": "ContextualAI/archangel_sft-dpo_pythia12-0b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5009 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6676 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.364 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5432 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4139 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5303 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia2-8b/7594c6a6-43af-49e5-a2b0-f41448c2e273.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia2-8b/7594c6a6-43af-49e5-a2b0-f41448c2e273.json new file mode 100644 index 0000000000000000000000000000000000000000..5fe4d230d625daeae4e1c346b9cd16d69043370f --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia2-8b/7594c6a6-43af-49e5-a2b0-f41448c2e273.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia2-8b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_pythia2-8b", + "id": "ContextualAI/archangel_sft-dpo_pythia2-8b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5286 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8073 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3355 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4473 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5135 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5501 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia6-9b/f422e6aa-a4b3-44b9-9694-450dc74b20eb.json b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia6-9b/f422e6aa-a4b3-44b9-9694-450dc74b20eb.json new file mode 100644 index 0000000000000000000000000000000000000000..5424d21b895aa1f4c800e50c48278973a0a8ac2d --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-dpo_pythia6-9b/f422e6aa-a4b3-44b9-9694-450dc74b20eb.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-dpo_pythia6-9b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-dpo_pythia6-9b", + "id": "ContextualAI/archangel_sft-dpo_pythia6-9b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5263 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7486 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3421 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5176 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4847 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.551 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_llama13b/de2ae5b9-3b11-497f-b9a3-03549bbdba97.json b/data/rewardbench/ContextualAI/archangel_sft-kto_llama13b/de2ae5b9-3b11-497f-b9a3-03549bbdba97.json new file mode 100644 index 0000000000000000000000000000000000000000..da7fc532bfdd7239ad35a9b17733a480343116ec --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_llama13b/de2ae5b9-3b11-497f-b9a3-03549bbdba97.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_llama13b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_llama13b", + "id": "ContextualAI/archangel_sft-kto_llama13b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5952 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8408 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3772 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4649 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7077 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.576 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_llama30b/a8f1f443-ef50-441e-9416-aa91d18463cf.json b/data/rewardbench/ContextualAI/archangel_sft-kto_llama30b/a8f1f443-ef50-441e-9416-aa91d18463cf.json new file mode 100644 index 0000000000000000000000000000000000000000..96afe6ff0f41d9fd99f292cd02e563a574bbb4d1 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_llama30b/a8f1f443-ef50-441e-9416-aa91d18463cf.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_llama30b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_llama30b", + "id": "ContextualAI/archangel_sft-kto_llama30b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5901 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8436 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4057 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6054 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5075 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5862 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_llama7b/59590dfb-9525-4999-95d8-cee22980f19f.json b/data/rewardbench/ContextualAI/archangel_sft-kto_llama7b/59590dfb-9525-4999-95d8-cee22980f19f.json new file mode 100644 index 0000000000000000000000000000000000000000..437b7a723f20b01ce7c9bd911dfb8c22e8c199aa --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_llama7b/59590dfb-9525-4999-95d8-cee22980f19f.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_llama7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_llama7b", + "id": "ContextualAI/archangel_sft-kto_llama7b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5388 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5587 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4364 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4568 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6941 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5575 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_pythia1-4b/8be424e9-c7f8-4aae-8776-e9af5d8b9407.json b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia1-4b/8be424e9-c7f8-4aae-8776-e9af5d8b9407.json new file mode 100644 index 0000000000000000000000000000000000000000..eb6ca16bc72bea240c299b776c06b1aaae998792 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia1-4b/8be424e9-c7f8-4aae-8776-e9af5d8b9407.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_pythia1-4b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_pythia1-4b", + "id": "ContextualAI/archangel_sft-kto_pythia1-4b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5581 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6844 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3794 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5257 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6447 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5546 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_pythia12-0b/53e58eba-9615-4f44-bccb-b2232fad9c14.json b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia12-0b/53e58eba-9615-4f44-bccb-b2232fad9c14.json new file mode 100644 index 0000000000000000000000000000000000000000..a8d4dbd2a53d930677a57d6d72cd682750fb688a --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia12-0b/53e58eba-9615-4f44-bccb-b2232fad9c14.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_pythia12-0b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_pythia12-0b", + "id": "ContextualAI/archangel_sft-kto_pythia12-0b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5053 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7486 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3618 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4127 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.55 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_pythia2-8b/30e82ceb-6784-48c3-8e1d-8f2bddd86de9.json b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia2-8b/30e82ceb-6784-48c3-8e1d-8f2bddd86de9.json new file mode 100644 index 0000000000000000000000000000000000000000..911b4674dc4d093a23895abc62ecbdfe868cee8a --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia2-8b/30e82ceb-6784-48c3-8e1d-8f2bddd86de9.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_pythia2-8b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_pythia2-8b", + "id": "ContextualAI/archangel_sft-kto_pythia2-8b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5497 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.757 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3421 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4743 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6216 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.557 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ContextualAI/archangel_sft-kto_pythia6-9b/84d869cc-9f0f-4171-8735-d5277efb843d.json b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia6-9b/84d869cc-9f0f-4171-8735-d5277efb843d.json new file mode 100644 index 0000000000000000000000000000000000000000..58f4356acadd4140662a64e3a9a21fd189bc84b7 --- /dev/null +++ b/data/rewardbench/ContextualAI/archangel_sft-kto_pythia6-9b/84d869cc-9f0f-4171-8735-d5277efb843d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ContextualAI_archangel_sft-kto_pythia6-9b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ContextualAI/archangel_sft-kto_pythia6-9b", + "id": "ContextualAI/archangel_sft-kto_pythia6-9b", + "developer": "ContextualAI", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5561 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7765 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3618 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5365 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5415 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5723 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Databricks-Mosaic-Research/PGRM/12cac67d-a414-4286-9737-d016c9475f33.json b/data/rewardbench/Databricks-Mosaic-Research/PGRM/12cac67d-a414-4286-9737-d016c9475f33.json new file mode 100644 index 0000000000000000000000000000000000000000..637973faa02b7ac9e134b6049ae69c9d14a920b5 --- /dev/null +++ b/data/rewardbench/Databricks-Mosaic-Research/PGRM/12cac67d-a414-4286-9737-d016c9475f33.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Databricks-Mosaic-Research_PGRM/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Databricks-Mosaic-Research/PGRM", + "id": "Databricks-Mosaic-Research/PGRM", + "developer": "Databricks-Mosaic-Research", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8002 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7937 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7404 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9289 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9424 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8893 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/HFXM/RAMO-Llama3.1-8B/9ce47dd8-844d-4f92-b8fe-b7d074867d63.json b/data/rewardbench/HFXM/RAMO-Llama3.1-8B/9ce47dd8-844d-4f92-b8fe-b7d074867d63.json new file mode 100644 index 0000000000000000000000000000000000000000..72ec13a77e12b6d5071681ec3b736f9e666f7823 --- /dev/null +++ b/data/rewardbench/HFXM/RAMO-Llama3.1-8B/9ce47dd8-844d-4f92-b8fe-b7d074867d63.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/HFXM_RAMO-Llama3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "HFXM/RAMO-Llama3.1-8B", + "id": "HFXM/RAMO-Llama3.1-8B", + "developer": "HFXM", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6917 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6547 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5628 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9756 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9071 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6752 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/HuggingFaceH4/starchat2-15b-v0.1/bccca7e1-0396-4197-acb7-ebab5b79e9ec.json b/data/rewardbench/HuggingFaceH4/starchat2-15b-v0.1/bccca7e1-0396-4197-acb7-ebab5b79e9ec.json new file mode 100644 index 0000000000000000000000000000000000000000..907769cf6a0ef7bc553ef4fc8703050050b820a1 --- /dev/null +++ b/data/rewardbench/HuggingFaceH4/starchat2-15b-v0.1/bccca7e1-0396-4197-acb7-ebab5b79e9ec.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/HuggingFaceH4_starchat2-15b-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "HuggingFaceH4/starchat2-15b-v0.1", + "id": "HuggingFaceH4/starchat2-15b-v0.1", + "developer": "HuggingFaceH4", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7322 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9385 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5548 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7095 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8159 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5525 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/HuggingFaceH4/zephyr-7b-alpha/5d2ed8a7-0410-47b5-b73c-a73797bbaa36.json b/data/rewardbench/HuggingFaceH4/zephyr-7b-alpha/5d2ed8a7-0410-47b5-b73c-a73797bbaa36.json new file mode 100644 index 0000000000000000000000000000000000000000..761c0061555d601c09400b5723eb77651481d2e8 --- /dev/null +++ b/data/rewardbench/HuggingFaceH4/zephyr-7b-alpha/5d2ed8a7-0410-47b5-b73c-a73797bbaa36.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/HuggingFaceH4_zephyr-7b-alpha/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "HuggingFaceH4/zephyr-7b-alpha", + "id": "HuggingFaceH4/zephyr-7b-alpha", + "developer": "HuggingFaceH4", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7392 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9162 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.625 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7662 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7514 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5353 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/HuggingFaceH4/zephyr-7b-beta/affa7458-5059-49c7-9c34-0303b35e3c55.json b/data/rewardbench/HuggingFaceH4/zephyr-7b-beta/affa7458-5059-49c7-9c34-0303b35e3c55.json new file mode 100644 index 0000000000000000000000000000000000000000..cfdfeee715d2c5189c2521269a1ca3fa9c4e2f2a --- /dev/null +++ b/data/rewardbench/HuggingFaceH4/zephyr-7b-beta/affa7458-5059-49c7-9c34-0303b35e3c55.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/HuggingFaceH4_zephyr-7b-beta/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "HuggingFaceH4/zephyr-7b-beta", + "id": "HuggingFaceH4/zephyr-7b-beta", + "developer": "HuggingFaceH4", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7281 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6272 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6568 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7789 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5216 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/HuggingFaceH4/zephyr-7b-gemma-v0.1/3bcbf294-c55d-45d0-ad40-5e105c873fda.json b/data/rewardbench/HuggingFaceH4/zephyr-7b-gemma-v0.1/3bcbf294-c55d-45d0-ad40-5e105c873fda.json new file mode 100644 index 0000000000000000000000000000000000000000..2b43d268f664f3317e178f44232d6926c1e0e67c --- /dev/null +++ b/data/rewardbench/HuggingFaceH4/zephyr-7b-gemma-v0.1/3bcbf294-c55d-45d0-ad40-5e105c873fda.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/HuggingFaceH4_zephyr-7b-gemma-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "HuggingFaceH4/zephyr-7b-gemma-v0.1", + "id": "HuggingFaceH4/zephyr-7b-gemma-v0.1", + "developer": "HuggingFaceH4", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6758 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4956 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5824 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7463 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5171 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/IDEA-CCNL/Ziya-LLaMA-7B-Reward/cdb0c3e4-0889-4050-8e5c-16c9a5b6c07d.json b/data/rewardbench/IDEA-CCNL/Ziya-LLaMA-7B-Reward/cdb0c3e4-0889-4050-8e5c-16c9a5b6c07d.json new file mode 100644 index 0000000000000000000000000000000000000000..b29a7e782d346c9db172ac93f235f85cef313722 --- /dev/null +++ b/data/rewardbench/IDEA-CCNL/Ziya-LLaMA-7B-Reward/cdb0c3e4-0889-4050-8e5c-16c9a5b6c07d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/IDEA-CCNL_Ziya-LLaMA-7B-Reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "IDEA-CCNL/Ziya-LLaMA-7B-Reward", + "id": "IDEA-CCNL/Ziya-LLaMA-7B-Reward", + "developer": "IDEA-CCNL", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6378 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8687 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4605 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6405 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5775 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6461 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/LxzGordon/URM-LLaMa-3-8B/0715b343-342c-4986-a6f1-fe69e5012cd9.json b/data/rewardbench/LxzGordon/URM-LLaMa-3-8B/0715b343-342c-4986-a6f1-fe69e5012cd9.json new file mode 100644 index 0000000000000000000000000000000000000000..6900eabcaa88d7d34a3b038b9387b3be5503fa56 --- /dev/null +++ b/data/rewardbench/LxzGordon/URM-LLaMa-3-8B/0715b343-342c-4986-a6f1-fe69e5012cd9.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/LxzGordon_URM-LLaMa-3-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "LxzGordon/URM-LLaMa-3-8B", + "id": "LxzGordon/URM-LLaMa-3-8B", + "developer": "LxzGordon", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8991 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7873 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8824 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9574 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/55b02489-273d-47e2-8026-502499731d8f.json b/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/55b02489-273d-47e2-8026-502499731d8f.json new file mode 100644 index 0000000000000000000000000000000000000000..9c163e8e10ff7935f69b725204f20204997a1fdc --- /dev/null +++ b/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/55b02489-273d-47e2-8026-502499731d8f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/LxzGordon_URM-LLaMa-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "LxzGordon/URM-LLaMa-3.1-8B", + "id": "LxzGordon/URM-LLaMa-3.1-8B", + "developer": "LxzGordon", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7394 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6884 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.45 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6393 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9758 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7653 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/6440ddc7-7cbf-4c85-865b-9b8bac6dc620.json b/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/6440ddc7-7cbf-4c85-865b-9b8bac6dc620.json new file mode 100644 index 0000000000000000000000000000000000000000..c5d5f5a2cddeace3e9b455f1316af392dc459bd8 --- /dev/null +++ b/data/rewardbench/LxzGordon/URM-LLaMa-3.1-8B/6440ddc7-7cbf-4c85-865b-9b8bac6dc620.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/LxzGordon_URM-LLaMa-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "LxzGordon/URM-LLaMa-3.1-8B", + "id": "LxzGordon/URM-LLaMa-3.1-8B", + "developer": "LxzGordon", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9294 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8816 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9108 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9698 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NCSOFT/Llama-3-OffsetBias-8B/9180bd80-5082-40ed-a5a9-26975cc509d2.json b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-8B/9180bd80-5082-40ed-a5a9-26975cc509d2.json new file mode 100644 index 0000000000000000000000000000000000000000..285d7b473f6f9343e0576f783e131d249c9ab3f5 --- /dev/null +++ b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-8B/9180bd80-5082-40ed-a5a9-26975cc509d2.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/NCSOFT_Llama-3-OffsetBias-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NCSOFT/Llama-3-OffsetBias-8B", + "id": "NCSOFT/Llama-3-OffsetBias-8B", + "developer": "NCSOFT", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8397 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9246 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8026 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8676 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7639 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/70d7ce2e-f685-4d17-a484-f1cfdc92bebd.json b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/70d7ce2e-f685-4d17-a484-f1cfdc92bebd.json new file mode 100644 index 0000000000000000000000000000000000000000..37a1ed3ec06eebfbdc73eb43968595f636f76158 --- /dev/null +++ b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/70d7ce2e-f685-4d17-a484-f1cfdc92bebd.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/NCSOFT_Llama-3-OffsetBias-RM-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NCSOFT/Llama-3-OffsetBias-RM-8B", + "id": "NCSOFT/Llama-3-OffsetBias-RM-8B", + "developer": "NCSOFT", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8942 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9721 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.818 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8676 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9192 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/87845f3c-8a0c-46a9-bbe0-829f1d8fc4ed.json b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/87845f3c-8a0c-46a9-bbe0-829f1d8fc4ed.json new file mode 100644 index 0000000000000000000000000000000000000000..98b47d0662851b25a4d784b4eb29a1b6a85ccb97 --- /dev/null +++ b/data/rewardbench/NCSOFT/Llama-3-OffsetBias-RM-8B/87845f3c-8a0c-46a9-bbe0-829f1d8fc4ed.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/NCSOFT_Llama-3-OffsetBias-RM-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NCSOFT/Llama-3-OffsetBias-RM-8B", + "id": "NCSOFT/Llama-3-OffsetBias-RM-8B", + "developer": "NCSOFT", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.648 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5191 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6786 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Nexusflow/Starling-RM-34B/4d47bb4c-5175-49ff-99aa-277b9c323791.json b/data/rewardbench/Nexusflow/Starling-RM-34B/4d47bb4c-5175-49ff-99aa-277b9c323791.json new file mode 100644 index 0000000000000000000000000000000000000000..b9e6e7d768da7cb388aef29e8e97560b1dadd02b --- /dev/null +++ b/data/rewardbench/Nexusflow/Starling-RM-34B/4d47bb4c-5175-49ff-99aa-277b9c323791.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Nexusflow_Starling-RM-34B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Nexusflow/Starling-RM-34B", + "id": "Nexusflow/Starling-RM-34B", + "developer": "Nexusflow", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8133 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5724 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.877 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8845 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7137 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Nexusflow/Starling-RM-34B/bb5614f7-9a53-47be-9b8f-1431ebde7d93.json b/data/rewardbench/Nexusflow/Starling-RM-34B/bb5614f7-9a53-47be-9b8f-1431ebde7d93.json new file mode 100644 index 0000000000000000000000000000000000000000..5f76dad4c8ef1532206227b6116e646c3aa3ca2d --- /dev/null +++ b/data/rewardbench/Nexusflow/Starling-RM-34B/bb5614f7-9a53-47be-9b8f-1431ebde7d93.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Nexusflow_Starling-RM-34B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Nexusflow/Starling-RM-34B", + "id": "Nexusflow/Starling-RM-34B", + "developer": "Nexusflow", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4553 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4589 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3187 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7556 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4808 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1004 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NousResearch/Hermes-3-Llama-3.1-70B/e00a773d-6df0-4c34-8a00-bd2426b03a9e.json b/data/rewardbench/NousResearch/Hermes-3-Llama-3.1-70B/e00a773d-6df0-4c34-8a00-bd2426b03a9e.json new file mode 100644 index 0000000000000000000000000000000000000000..e7cc3cceb12f5fe1bd3e9dd461a9cd51d7d1e05d --- /dev/null +++ b/data/rewardbench/NousResearch/Hermes-3-Llama-3.1-70B/e00a773d-6df0-4c34-8a00-bd2426b03a9e.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/NousResearch_Hermes-3-Llama-3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NousResearch/Hermes-3-Llama-3.1-70B", + "id": "NousResearch/Hermes-3-Llama-3.1-70B", + "developer": "NousResearch", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7847 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9623 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5669 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.823 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7867 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NousResearch/Nous-Hermes-2-Mistral-7B-DPO/45adc2a4-b6cc-48e5-82ec-7a9d9b06f8f3.json b/data/rewardbench/NousResearch/Nous-Hermes-2-Mistral-7B-DPO/45adc2a4-b6cc-48e5-82ec-7a9d9b06f8f3.json new file mode 100644 index 0000000000000000000000000000000000000000..49ac58486858ea32822b78a6a19fd3ebaf228d5d --- /dev/null +++ b/data/rewardbench/NousResearch/Nous-Hermes-2-Mistral-7B-DPO/45adc2a4-b6cc-48e5-82ec-7a9d9b06f8f3.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/NousResearch_Nous-Hermes-2-Mistral-7B-DPO/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NousResearch/Nous-Hermes-2-Mistral-7B-DPO", + "id": "NousResearch/Nous-Hermes-2-Mistral-7B-DPO", + "developer": "NousResearch", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7481 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6053 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8243 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7375 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.555 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO/271d8498-4071-4b88-a084-ad6ee846884c.json b/data/rewardbench/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO/271d8498-4071-4b88-a084-ad6ee846884c.json new file mode 100644 index 0000000000000000000000000000000000000000..c95f33aff7e96d053c7e7d95a086d0c06518c790 --- /dev/null +++ b/data/rewardbench/NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO/271d8498-4071-4b88-a084-ad6ee846884c.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/NousResearch_Nous-Hermes-2-Mixtral-8x7B-DPO/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO", + "id": "NousResearch/Nous-Hermes-2-Mixtral-8x7B-DPO", + "developer": "NousResearch", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7138 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9162 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6053 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8149 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6126 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5266 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/06878ea7-2daa-45f0-8b50-9d5239ad27ee.json b/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/06878ea7-2daa-45f0-8b50-9d5239ad27ee.json new file mode 100644 index 0000000000000000000000000000000000000000..77ee183f33863a2d2b2e8c5daebde53a80aca616 --- /dev/null +++ b/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/06878ea7-2daa-45f0-8b50-9d5239ad27ee.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/OpenAssistant_oasst-rm-2-pythia-6.9b-epoch-1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1", + "id": "OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2653 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3979 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.377 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3289 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1535 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.047 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/e5f04b5d-f48f-400e-ae5e-a7b8b46773f9.json b/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/e5f04b5d-f48f-400e-ae5e-a7b8b46773f9.json new file mode 100644 index 0000000000000000000000000000000000000000..9e59bdd829ddf3834e7f74f1f6132f3bd1a8aa12 --- /dev/null +++ b/data/rewardbench/OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1/e5f04b5d-f48f-400e-ae5e-a7b8b46773f9.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/OpenAssistant_oasst-rm-2-pythia-6.9b-epoch-1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1", + "id": "OpenAssistant/oasst-rm-2-pythia-6.9b-epoch-1", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.615 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9246 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3728 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5855 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6801 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/bfa12e46-28fe-44c4-829b-8bfec4ecf860.json b/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/bfa12e46-28fe-44c4-829b-8bfec4ecf860.json new file mode 100644 index 0000000000000000000000000000000000000000..b601b52738e1375449328e245bb69d0f1a31a443 --- /dev/null +++ b/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/bfa12e46-28fe-44c4-829b-8bfec4ecf860.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/OpenAssistant_oasst-rm-2.1-pythia-1.4b-epoch-2.5/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5", + "id": "OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6901 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8855 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4868 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6311 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7752 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6533 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/f424f1ef-959c-4c7a-b72c-cf5addb266dc.json b/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/f424f1ef-959c-4c7a-b72c-cf5addb266dc.json new file mode 100644 index 0000000000000000000000000000000000000000..b7d227cb3d1c31ca7663cd606bf14759e67d0b52 --- /dev/null +++ b/data/rewardbench/OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5/f424f1ef-959c-4c7a-b72c-cf5addb266dc.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/OpenAssistant_oasst-rm-2.1-pythia-1.4b-epoch-2.5/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5", + "id": "OpenAssistant/oasst-rm-2.1-pythia-1.4b-epoch-2.5", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2648 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3179 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3934 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3244 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2707 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0198 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/4a45257e-7a2f-46ae-b2f0-aac7c912b7b6.json b/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/4a45257e-7a2f-46ae-b2f0-aac7c912b7b6.json new file mode 100644 index 0000000000000000000000000000000000000000..944052b50aa735b8be6bd82e3b349f6c4469232d --- /dev/null +++ b/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/4a45257e-7a2f-46ae-b2f0-aac7c912b7b6.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/OpenAssistant_reward-model-deberta-v3-large-v2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/reward-model-deberta-v3-large-v2", + "id": "OpenAssistant/reward-model-deberta-v3-large-v2", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6126 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8939 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4518 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7338 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3855 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5836 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/b0124154-e9f5-42d1-ad59-3d519a06c439.json b/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/b0124154-e9f5-42d1-ad59-3d519a06c439.json new file mode 100644 index 0000000000000000000000000000000000000000..af9ea75cd8aa3516a4f01073ffaaaf73ad51e417 --- /dev/null +++ b/data/rewardbench/OpenAssistant/reward-model-deberta-v3-large-v2/b0124154-e9f5-42d1-ad59-3d519a06c439.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/OpenAssistant_reward-model-deberta-v3-large-v2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "OpenAssistant/reward-model-deberta-v3-large-v2", + "id": "OpenAssistant/reward-model-deberta-v3-large-v2", + "developer": "OpenAssistant", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.32 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3853 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2687 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5027 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2768 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.12 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/002e0f6d-4ec2-407b-9be0-450839f6d0f5.json b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/002e0f6d-4ec2-407b-9be0-450839f6d0f5.json new file mode 100644 index 0000000000000000000000000000000000000000..079b243095d0b406e0342df87fa142f8e8624520 --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/002e0f6d-4ec2-407b-9be0-450839f6d0f5.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/PKU-Alignment_beaver-7b-v1.0-cost/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v1.0-cost", + "id": "PKU-Alignment/beaver-7b-v1.0-cost", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5798 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6173 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4232 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7351 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5482 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.57 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/b8b7fea1-0017-4a2c-8612-6f844043dc52.json b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/b8b7fea1-0017-4a2c-8612-6f844043dc52.json new file mode 100644 index 0000000000000000000000000000000000000000..5f3a1859a31b40c6091a050463d0091a65bfca83 --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-cost/b8b7fea1-0017-4a2c-8612-6f844043dc52.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/PKU-Alignment_beaver-7b-v1.0-cost/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v1.0-cost", + "id": "PKU-Alignment/beaver-7b-v1.0-cost", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3332 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2313 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3989 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7589 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2939 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b4c39967-b02b-4997-a454-6b635c85b55f.json b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b4c39967-b02b-4997-a454-6b635c85b55f.json new file mode 100644 index 0000000000000000000000000000000000000000..6b2193e0b3b15eccd177924f2bd92cabfcbef979 --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b4c39967-b02b-4997-a454-6b635c85b55f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/PKU-Alignment_beaver-7b-v1.0-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v1.0-reward", + "id": "PKU-Alignment/beaver-7b-v1.0-reward", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1606 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2105 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2938 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0646 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b6aeae08-a833-4b3a-9f3c-8972937c855b.json b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b6aeae08-a833-4b3a-9f3c-8972937c855b.json new file mode 100644 index 0000000000000000000000000000000000000000..d92cc53ba411b85f335e8c235d11f50aec4fe77c --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v1.0-reward/b6aeae08-a833-4b3a-9f3c-8972937c855b.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/PKU-Alignment_beaver-7b-v1.0-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v1.0-reward", + "id": "PKU-Alignment/beaver-7b-v1.0-reward", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4727 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8184 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2873 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.346 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5993 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/015cac72-546c-47b4-bf0e-a35ad74b3cf0.json b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/015cac72-546c-47b4-bf0e-a35ad74b3cf0.json new file mode 100644 index 0000000000000000000000000000000000000000..badd0e14d9b868adec5f4e5291c8b910be9fbbad --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/015cac72-546c-47b4-bf0e-a35ad74b3cf0.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/PKU-Alignment_beaver-7b-v2.0-cost/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v2.0-cost", + "id": "PKU-Alignment/beaver-7b-v2.0-cost", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3326 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3789 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.275 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3333 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7356 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2828 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/e4bbaeda-58c5-4703-9de9-272060c6340d.json b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/e4bbaeda-58c5-4703-9de9-272060c6340d.json new file mode 100644 index 0000000000000000000000000000000000000000..b0288d1a8f70da86f4068826e48f005f9fe9e69b --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-cost/e4bbaeda-58c5-4703-9de9-272060c6340d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/PKU-Alignment_beaver-7b-v2.0-cost/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v2.0-cost", + "id": "PKU-Alignment/beaver-7b-v2.0-cost", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5957 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5726 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4561 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7608 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6211 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5397 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/0830e0e3-aea1-4c6d-8db8-7d4d7b43efcd.json b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/0830e0e3-aea1-4c6d-8db8-7d4d7b43efcd.json new file mode 100644 index 0000000000000000000000000000000000000000..71a95e99e0dfffcf1e814b38851af3f06bb9b649 --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/0830e0e3-aea1-4c6d-8db8-7d4d7b43efcd.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/PKU-Alignment_beaver-7b-v2.0-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v2.0-reward", + "id": "PKU-Alignment/beaver-7b-v2.0-reward", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6366 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8994 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.364 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6041 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6887 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6171 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/ae95c3e6-33a4-4d84-b769-0dbd71813de6.json b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/ae95c3e6-33a4-4d84-b769-0dbd71813de6.json new file mode 100644 index 0000000000000000000000000000000000000000..3729b377a241d102de31941e68e2d11feeb62dca --- /dev/null +++ b/data/rewardbench/PKU-Alignment/beaver-7b-v2.0-reward/ae95c3e6-33a4-4d84-b769-0dbd71813de6.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/PKU-Alignment_beaver-7b-v2.0-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PKU-Alignment/beaver-7b-v2.0-reward", + "id": "PKU-Alignment/beaver-7b-v2.0-reward", + "developer": "PKU-Alignment", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2544 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2168 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2562 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3825 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3156 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2606 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0944 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/PoLL/gpt-3.5-turbo-0125_claude-3-sonnet-2024022.../8c9f8d4f-7fc5-426b-892a-2ca6568358af.json b/data/rewardbench/PoLL/gpt-3.5-turbo-0125_claude-3-sonnet-2024022.../8c9f8d4f-7fc5-426b-892a-2ca6568358af.json new file mode 100644 index 0000000000000000000000000000000000000000..eaf1e196092d45316f0b702624534ed88d1f845d --- /dev/null +++ b/data/rewardbench/PoLL/gpt-3.5-turbo-0125_claude-3-sonnet-2024022.../8c9f8d4f-7fc5-426b-892a-2ca6568358af.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/PoLL_gpt-3.5-turbo-0125_claude-3-sonnet-2024022.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "PoLL/gpt-3.5-turbo-0125_claude-3-sonnet-2024022...", + "id": "PoLL/gpt-3.5-turbo-0125_claude-3-sonnet-2024022...", + "developer": "PoLL", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7578 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5406 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8034 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7346 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-0.5B-Chat/c3cc3da5-7b75-409b-9b9a-d8f5604248af.json b/data/rewardbench/Qwen/Qwen1.5-0.5B-Chat/c3cc3da5-7b75-409b-9b9a-d8f5604248af.json new file mode 100644 index 0000000000000000000000000000000000000000..b64ae0dd9d2ee5dc8afee12eaa4e9264b7ce0f93 --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-0.5B-Chat/c3cc3da5-7b75-409b-9b9a-d8f5604248af.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-0.5B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-0.5B-Chat", + "id": "Qwen/Qwen1.5-0.5B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5298 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3547 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6294 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5703 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5984 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4629 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-1.8B-Chat/0c73110e-431c-46a9-bcf5-6177582a04ae.json b/data/rewardbench/Qwen/Qwen1.5-1.8B-Chat/0c73110e-431c-46a9-bcf5-6177582a04ae.json new file mode 100644 index 0000000000000000000000000000000000000000..165a888c58bef567066bfad0a4acceb6b9e77062 --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-1.8B-Chat/0c73110e-431c-46a9-bcf5-6177582a04ae.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-1.8B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-1.8B-Chat", + "id": "Qwen/Qwen1.5-1.8B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.589 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5615 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6031 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4838 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7793 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4453 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-14B-Chat/5d11aefa-449c-465d-80bb-21dd7978f6d6.json b/data/rewardbench/Qwen/Qwen1.5-14B-Chat/5d11aefa-449c-465d-80bb-21dd7978f6d6.json new file mode 100644 index 0000000000000000000000000000000000000000..02ac20c517f9f550716908586279006aeb57e7e5 --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-14B-Chat/5d11aefa-449c-465d-80bb-21dd7978f6d6.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-14B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-14B-Chat", + "id": "Qwen/Qwen1.5-14B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6864 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5726 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7018 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7122 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8961 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4123 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-4B-Chat/d73bbe54-3001-4c6d-a248-1b3ad227d236.json b/data/rewardbench/Qwen/Qwen1.5-4B-Chat/d73bbe54-3001-4c6d-a248-1b3ad227d236.json new file mode 100644 index 0000000000000000000000000000000000000000..acd80539aa6ea00aedc68db0476bf31cc4274356 --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-4B-Chat/d73bbe54-3001-4c6d-a248-1b3ad227d236.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-4B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-4B-Chat", + "id": "Qwen/Qwen1.5-4B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5477 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3883 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6272 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5568 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6689 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.447 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-72B-Chat/ef33b6b1-bf7a-4c7c-bf45-a8ee4be55262.json b/data/rewardbench/Qwen/Qwen1.5-72B-Chat/ef33b6b1-bf7a-4c7c-bf45-a8ee4be55262.json new file mode 100644 index 0000000000000000000000000000000000000000..2e1f53bde61b5e611e59411daa2c326afe231a5d --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-72B-Chat/ef33b6b1-bf7a-4c7c-bf45-a8ee4be55262.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-72B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-72B-Chat", + "id": "Qwen/Qwen1.5-72B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6723 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6229 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6601 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8554 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4226 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-7B-Chat/a153f6d7-4041-4ce9-a590-507bd07af009.json b/data/rewardbench/Qwen/Qwen1.5-7B-Chat/a153f6d7-4041-4ce9-a590-507bd07af009.json new file mode 100644 index 0000000000000000000000000000000000000000..0e9a79589b403ad54f2108503f8a9d2791bf327e --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-7B-Chat/a153f6d7-4041-4ce9-a590-507bd07af009.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-7B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-7B-Chat", + "id": "Qwen/Qwen1.5-7B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.675 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5363 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6908 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6919 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9041 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4288 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/Qwen1.5-MoE-A2.7B-Chat/5bca682d-69ca-4705-855e-c8cafebbe9a3.json b/data/rewardbench/Qwen/Qwen1.5-MoE-A2.7B-Chat/5bca682d-69ca-4705-855e-c8cafebbe9a3.json new file mode 100644 index 0000000000000000000000000000000000000000..dee3b1115935f1e471adc6aa0a3fa24020d0136d --- /dev/null +++ b/data/rewardbench/Qwen/Qwen1.5-MoE-A2.7B-Chat/5bca682d-69ca-4705-855e-c8cafebbe9a3.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Qwen_Qwen1.5-MoE-A2.7B-Chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/Qwen1.5-MoE-A2.7B-Chat", + "id": "Qwen/Qwen1.5-MoE-A2.7B-Chat", + "developer": "Qwen", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6644 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7291 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6316 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.774 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4536 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Qwen/WorldPM-72B/40af66d8-99e3-412a-b7d6-3c44f7d57411.json b/data/rewardbench/Qwen/WorldPM-72B/40af66d8-99e3-412a-b7d6-3c44f7d57411.json new file mode 100644 index 0000000000000000000000000000000000000000..3834704849966cc5ad37ec3b428a376a0f94720a --- /dev/null +++ b/data/rewardbench/Qwen/WorldPM-72B/40af66d8-99e3-412a-b7d6-3c44f7d57411.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Qwen_WorldPM-72B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Qwen/WorldPM-72B", + "id": "Qwen/WorldPM-72B", + "developer": "Qwen", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6333 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8533 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9172 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3535 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-32B/80df2f8b-1716-4eeb-ba8b-90e86a4cd127.json b/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-32B/80df2f8b-1716-4eeb-ba8b-90e86a4cd127.json new file mode 100644 index 0000000000000000000000000000000000000000..992f24120d35794ed78e3b38474cea6b14861a3c --- /dev/null +++ b/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-32B/80df2f8b-1716-4eeb-ba8b-90e86a4cd127.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/R-I-S-E_RISE-Judge-Qwen2.5-32B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "R-I-S-E/RISE-Judge-Qwen2.5-32B", + "id": "R-I-S-E/RISE-Judge-Qwen2.5-32B", + "developer": "R-I-S-E", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9266 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8333 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9189 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9877 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-7B/1d0857da-e330-4ff6-9c26-a40611411825.json b/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-7B/1d0857da-e330-4ff6-9c26-a40611411825.json new file mode 100644 index 0000000000000000000000000000000000000000..4b66f9a95ff5b0f8790d6a42fb59401e6a7cd8f9 --- /dev/null +++ b/data/rewardbench/R-I-S-E/RISE-Judge-Qwen2.5-7B/1d0857da-e330-4ff6-9c26-a40611411825.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/R-I-S-E_RISE-Judge-Qwen2.5-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "R-I-S-E/RISE-Judge-Qwen2.5-7B", + "id": "R-I-S-E/RISE-Judge-Qwen2.5-7B", + "developer": "R-I-S-E", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8819 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7654 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8797 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9608 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/1b82e3b1-7a22-477a-9e49-85c4800395d8.json b/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/1b82e3b1-7a22-477a-9e49-85c4800395d8.json new file mode 100644 index 0000000000000000000000000000000000000000..24cfb9985b2edeaf964405998ec8888a7c73a6f8 --- /dev/null +++ b/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/1b82e3b1-7a22-477a-9e49-85c4800395d8.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/RLHFlow_ArmoRM-Llama3-8B-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "RLHFlow/ArmoRM-Llama3-8B-v0.1", + "id": "RLHFlow/ArmoRM-Llama3-8B-v0.1", + "developer": "RLHFlow", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6646 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6568 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7657 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6629 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/e041806f-c4c2-4ae2-aaf5-ae1245955006.json b/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/e041806f-c4c2-4ae2-aaf5-ae1245955006.json new file mode 100644 index 0000000000000000000000000000000000000000..92616465ff0af2e2fa178300897e4900e2cc2cfa --- /dev/null +++ b/data/rewardbench/RLHFlow/ArmoRM-Llama3-8B-v0.1/e041806f-c4c2-4ae2-aaf5-ae1245955006.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/RLHFlow_ArmoRM-Llama3-8B-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "RLHFlow/ArmoRM-Llama3-8B-v0.1", + "id": "RLHFlow/ArmoRM-Llama3-8B-v0.1", + "developer": "RLHFlow", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.886 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7675 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9054 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9735 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7429 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/RLHFlow/LLaMA3-iterative-DPO-final/be5f9fb3-71de-4229-9c95-925472fd348c.json b/data/rewardbench/RLHFlow/LLaMA3-iterative-DPO-final/be5f9fb3-71de-4229-9c95-925472fd348c.json new file mode 100644 index 0000000000000000000000000000000000000000..06d6049ff8e9284e31ca341eee304600bd7dfb96 --- /dev/null +++ b/data/rewardbench/RLHFlow/LLaMA3-iterative-DPO-final/be5f9fb3-71de-4229-9c95-925472fd348c.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/RLHFlow_LLaMA3-iterative-DPO-final/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "RLHFlow/LLaMA3-iterative-DPO-final", + "id": "RLHFlow/LLaMA3-iterative-DPO-final", + "developer": "RLHFlow", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6783 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.838 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5921 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7865 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6161 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4392 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/RLHFlow/RewardModel-Mistral-7B-for-DPA-v1/6456844f-7a8b-4859-a4d2-fda442c2a857.json b/data/rewardbench/RLHFlow/RewardModel-Mistral-7B-for-DPA-v1/6456844f-7a8b-4859-a4d2-fda442c2a857.json new file mode 100644 index 0000000000000000000000000000000000000000..b0235d3725f2baf9528ba957ce9c428d348a1195 --- /dev/null +++ b/data/rewardbench/RLHFlow/RewardModel-Mistral-7B-for-DPA-v1/6456844f-7a8b-4859-a4d2-fda442c2a857.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/RLHFlow_RewardModel-Mistral-7B-for-DPA-v1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "RLHFlow/RewardModel-Mistral-7B-for-DPA-v1", + "id": "RLHFlow/RewardModel-Mistral-7B-for-DPA-v1", + "developer": "RLHFlow", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6633 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8799 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4978 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7068 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5971 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6068 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/RLHFlow/pair-preference-model-LLaMA3-8B/f5daa71d-3f45-445a-a413-75049fb534db.json b/data/rewardbench/RLHFlow/pair-preference-model-LLaMA3-8B/f5daa71d-3f45-445a-a413-75049fb534db.json new file mode 100644 index 0000000000000000000000000000000000000000..c42e8d8c222779cc9c9fec9cbfd7cd5bb33eaaeb --- /dev/null +++ b/data/rewardbench/RLHFlow/pair-preference-model-LLaMA3-8B/f5daa71d-3f45-445a-a413-75049fb534db.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/RLHFlow_pair-preference-model-LLaMA3-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "RLHFlow/pair-preference-model-LLaMA3-8B", + "id": "RLHFlow/pair-preference-model-LLaMA3-8B", + "developer": "RLHFlow", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8575 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9832 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6579 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8973 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9473 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7458 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-Gemma-2B-rewardmodel-ft/d224390f-7592-4400-8384-25875f81eb79.json b/data/rewardbench/Ray2333/GRM-Gemma-2B-rewardmodel-ft/d224390f-7592-4400-8384-25875f81eb79.json new file mode 100644 index 0000000000000000000000000000000000000000..98c803e332ab50b557e8a711db24fb2e145e0b47 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-Gemma-2B-rewardmodel-ft/d224390f-7592-4400-8384-25875f81eb79.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-Gemma-2B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-Gemma-2B-rewardmodel-ft", + "id": "Ray2333/GRM-Gemma-2B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8447 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8939 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7522 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8881 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-Gemma-2B-sftreg/d70e2b48-b3b0-43a8-b164-ec4bccf4897d.json b/data/rewardbench/Ray2333/GRM-Gemma-2B-sftreg/d70e2b48-b3b0-43a8-b164-ec4bccf4897d.json new file mode 100644 index 0000000000000000000000000000000000000000..f897051b25a1274a6197f8def27ed3722574d600 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-Gemma-2B-sftreg/d70e2b48-b3b0-43a8-b164-ec4bccf4897d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-Gemma-2B-sftreg/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-Gemma-2B-sftreg", + "id": "Ray2333/GRM-Gemma-2B-sftreg", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7451 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4868 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7932 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7684 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6983 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/2b3faa4e-a9b6-4db6-99f3-73b09376ae7b.json b/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/2b3faa4e-a9b6-4db6-99f3-73b09376ae7b.json new file mode 100644 index 0000000000000000000000000000000000000000..b11e14645aee3611e95982a18dd1172532fe2863 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/2b3faa4e-a9b6-4db6-99f3-73b09376ae7b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Ray2333_GRM-Llama3-8B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-Llama3-8B-rewardmodel-ft", + "id": "Ray2333/GRM-Llama3-8B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6766 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6274 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5847 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8929 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6824 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/c992751b-bccb-4a7a-b388-099ff8cd932b.json b/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/c992751b-bccb-4a7a-b388-099ff8cd932b.json new file mode 100644 index 0000000000000000000000000000000000000000..c0a48c824c909c2343aa53fbecf7c9e085112872 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-Llama3-8B-rewardmodel-ft/c992751b-bccb-4a7a-b388-099ff8cd932b.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-Llama3-8B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-Llama3-8B-rewardmodel-ft", + "id": "Ray2333/GRM-Llama3-8B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9154 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8618 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9081 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9362 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/1863fb2d-a620-43d2-9ca6-7162d4c3ff19.json b/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/1863fb2d-a620-43d2-9ca6-7162d4c3ff19.json new file mode 100644 index 0000000000000000000000000000000000000000..d141f9ea746551b33532abdc892940382e44a793 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/1863fb2d-a620-43d2-9ca6-7162d4c3ff19.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Ray2333_GRM-gemma2-2B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-gemma2-2B-rewardmodel-ft", + "id": "Ray2333/GRM-gemma2-2B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5966 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5305 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7455 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4788 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/bd040b58-30db-4c53-8f4e-3182c2d03d33.json b/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/bd040b58-30db-4c53-8f4e-3182c2d03d33.json new file mode 100644 index 0000000000000000000000000000000000000000..fc8e138d401b41ea7194594b443c01aad835ad9e --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-gemma2-2B-rewardmodel-ft/bd040b58-30db-4c53-8f4e-3182c2d03d33.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-gemma2-2B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-gemma2-2B-rewardmodel-ft", + "id": "Ray2333/GRM-gemma2-2B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8839 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9302 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7719 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9216 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.912 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-llama3-8B-distill/d4317c05-18b4-4385-b401-ead786b92ec5.json b/data/rewardbench/Ray2333/GRM-llama3-8B-distill/d4317c05-18b4-4385-b401-ead786b92ec5.json new file mode 100644 index 0000000000000000000000000000000000000000..b8adc92f7472fb4cc868b31a19c271b6b4a9f321 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-llama3-8B-distill/d4317c05-18b4-4385-b401-ead786b92ec5.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Ray2333_GRM-llama3-8B-distill/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-llama3-8B-distill", + "id": "Ray2333/GRM-llama3-8B-distill", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.589 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5874 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6727 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5743 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-llama3-8B-distill/ffc73ebe-bc67-490a-8976-e51ed38bf1d0.json b/data/rewardbench/Ray2333/GRM-llama3-8B-distill/ffc73ebe-bc67-490a-8976-e51ed38bf1d0.json new file mode 100644 index 0000000000000000000000000000000000000000..97bf7a41d0cf8f12b759519788e9f635be633564 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-llama3-8B-distill/ffc73ebe-bc67-490a-8976-e51ed38bf1d0.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-llama3-8B-distill/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-llama3-8B-distill", + "id": "Ray2333/GRM-llama3-8B-distill", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8464 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9832 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6842 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8676 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9133 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7209 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/8f91fd7e-b138-4274-a786-2ce0b3d820cd.json b/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/8f91fd7e-b138-4274-a786-2ce0b3d820cd.json new file mode 100644 index 0000000000000000000000000000000000000000..2f7a9884b6c5244ff2feb6e15085860e2c56af27 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/8f91fd7e-b138-4274-a786-2ce0b3d820cd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Ray2333_GRM-llama3-8B-sftreg/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-llama3-8B-sftreg", + "id": "Ray2333/GRM-llama3-8B-sftreg", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6089 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6189 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7867 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6828 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5981 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/cf311858-1550-4b31-bb92-13f053e60d29.json b/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/cf311858-1550-4b31-bb92-13f053e60d29.json new file mode 100644 index 0000000000000000000000000000000000000000..f769ac2d4ae239ff1777af8c22ffd48a0d97f24b --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-llama3-8B-sftreg/cf311858-1550-4b31-bb92-13f053e60d29.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-llama3-8B-sftreg/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-llama3-8B-sftreg", + "id": "Ray2333/GRM-llama3-8B-sftreg", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8542 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.986 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6776 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8919 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9229 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7309 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/GRM-llama3.2-3B-rewardmodel-ft/f036724f-48df-4f4e-b377-b8176f5e21cd.json b/data/rewardbench/Ray2333/GRM-llama3.2-3B-rewardmodel-ft/f036724f-48df-4f4e-b377-b8176f5e21cd.json new file mode 100644 index 0000000000000000000000000000000000000000..a679b496ee91b2a121dfcf612b5361e16513bbf6 --- /dev/null +++ b/data/rewardbench/Ray2333/GRM-llama3.2-3B-rewardmodel-ft/f036724f-48df-4f4e-b377-b8176f5e21cd.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_GRM-llama3.2-3B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/GRM-llama3.2-3B-rewardmodel-ft", + "id": "Ray2333/GRM-llama3.2-3B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9092 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9162 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8487 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.927 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.945 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-baseline/1e379181-6124-4ddd-9646-b0c7b53d77f6.json b/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-baseline/1e379181-6124-4ddd-9646-b0c7b53d77f6.json new file mode 100644 index 0000000000000000000000000000000000000000..3abf2002d7e1e8ab917037e043f26887e1c67ed5 --- /dev/null +++ b/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-baseline/1e379181-6124-4ddd-9646-b0c7b53d77f6.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_Gemma-2B-rewardmodel-baseline/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/Gemma-2B-rewardmodel-baseline", + "id": "Ray2333/Gemma-2B-rewardmodel-baseline", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.729 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9413 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4693 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7865 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7384 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6897 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-ft/2ef0ad6e-4f80-4c0e-98f3-c2443ceb05e8.json b/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-ft/2ef0ad6e-4f80-4c0e-98f3-c2443ceb05e8.json new file mode 100644 index 0000000000000000000000000000000000000000..52c037d287cbd6f3deef600b67f7b603cd106d55 --- /dev/null +++ b/data/rewardbench/Ray2333/Gemma-2B-rewardmodel-ft/2ef0ad6e-4f80-4c0e-98f3-c2443ceb05e8.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_Gemma-2B-rewardmodel-ft/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/Gemma-2B-rewardmodel-ft", + "id": "Ray2333/Gemma-2B-rewardmodel-ft", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8048 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7793 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7478 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8527 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8393 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Ray2333/reward-model-Mistral-7B-instruct-Unifie.../9e8a140c-5d91-494a-a0d8-69fbb2bda5e6.json b/data/rewardbench/Ray2333/reward-model-Mistral-7B-instruct-Unifie.../9e8a140c-5d91-494a-a0d8-69fbb2bda5e6.json new file mode 100644 index 0000000000000000000000000000000000000000..f9b3796b35a409b719e55caa5f1275dac17f6b9d --- /dev/null +++ b/data/rewardbench/Ray2333/reward-model-Mistral-7B-instruct-Unifie.../9e8a140c-5d91-494a-a0d8-69fbb2bda5e6.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Ray2333_reward-model-Mistral-7B-instruct-Unifie.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Ray2333/reward-model-Mistral-7B-instruct-Unifie...", + "id": "Ray2333/reward-model-Mistral-7B-instruct-Unifie...", + "developer": "Ray2333", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7661 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9777 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8527 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7389 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7434 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/SF-Foundation/TextEval-Llama3.1-70B/510a957c-02de-478a-834c-4783c1e61213.json b/data/rewardbench/SF-Foundation/TextEval-Llama3.1-70B/510a957c-02de-478a-834c-4783c1e61213.json new file mode 100644 index 0000000000000000000000000000000000000000..7162d796c8877b283f91237ca1aa660777678941 --- /dev/null +++ b/data/rewardbench/SF-Foundation/TextEval-Llama3.1-70B/510a957c-02de-478a-834c-4783c1e61213.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/SF-Foundation_TextEval-Llama3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "SF-Foundation/TextEval-Llama3.1-70B", + "id": "SF-Foundation/TextEval-Llama3.1-70B", + "developer": "SF-Foundation", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9348 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9413 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9013 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9324 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9641 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/SF-Foundation/TextEval-OffsetBias-12B/38c76802-dac0-42bf-9c85-3f06ddb39466.json b/data/rewardbench/SF-Foundation/TextEval-OffsetBias-12B/38c76802-dac0-42bf-9c85-3f06ddb39466.json new file mode 100644 index 0000000000000000000000000000000000000000..3fdfb4a9b3d0aca31d83fc667a8a192d458406a3 --- /dev/null +++ b/data/rewardbench/SF-Foundation/TextEval-OffsetBias-12B/38c76802-dac0-42bf-9c85-3f06ddb39466.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/SF-Foundation_TextEval-OffsetBias-12B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "SF-Foundation/TextEval-OffsetBias-12B", + "id": "SF-Foundation/TextEval-OffsetBias-12B", + "developer": "SF-Foundation", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9105 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.919 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8662 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9203 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9365 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Salesforce/SFR-LLaMa-3.1-70B-Judge-r/0ab8792d-784d-4ef8-b813-05299d5ab357.json b/data/rewardbench/Salesforce/SFR-LLaMa-3.1-70B-Judge-r/0ab8792d-784d-4ef8-b813-05299d5ab357.json new file mode 100644 index 0000000000000000000000000000000000000000..09160553ae924a5fcca219d863a2436ab00b4c25 --- /dev/null +++ b/data/rewardbench/Salesforce/SFR-LLaMa-3.1-70B-Judge-r/0ab8792d-784d-4ef8-b813-05299d5ab357.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Salesforce_SFR-LLaMa-3.1-70B-Judge-r/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Salesforce/SFR-LLaMa-3.1-70B-Judge-r", + "id": "Salesforce/SFR-LLaMa-3.1-70B-Judge-r", + "developer": "Salesforce", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9272 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8476 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9162 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9757 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Salesforce/SFR-LLaMa-3.1-8B-Judge-r/6fcc9215-e39e-4455-91ec-38f4c2575fcb.json b/data/rewardbench/Salesforce/SFR-LLaMa-3.1-8B-Judge-r/6fcc9215-e39e-4455-91ec-38f4c2575fcb.json new file mode 100644 index 0000000000000000000000000000000000000000..752ec7d139fde06cfb8f823f3558846106ca7a5a --- /dev/null +++ b/data/rewardbench/Salesforce/SFR-LLaMa-3.1-8B-Judge-r/6fcc9215-e39e-4455-91ec-38f4c2575fcb.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Salesforce_SFR-LLaMa-3.1-8B-Judge-r/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Salesforce/SFR-LLaMa-3.1-8B-Judge-r", + "id": "Salesforce/SFR-LLaMa-3.1-8B-Judge-r", + "developer": "Salesforce", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8865 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7774 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8622 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9513 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Salesforce/SFR-nemo-12B-Judge-r/0c530a7a-a4b8-4bb4-955d-3792063c8e22.json b/data/rewardbench/Salesforce/SFR-nemo-12B-Judge-r/0c530a7a-a4b8-4bb4-955d-3792063c8e22.json new file mode 100644 index 0000000000000000000000000000000000000000..446bf38cdd2e62887fc487135faf9fc0677eb25b --- /dev/null +++ b/data/rewardbench/Salesforce/SFR-nemo-12B-Judge-r/0c530a7a-a4b8-4bb4-955d-3792063c8e22.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Salesforce_SFR-nemo-12B-Judge-r/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Salesforce/SFR-nemo-12B-Judge-r", + "id": "Salesforce/SFR-nemo-12B-Judge-r", + "developer": "Salesforce", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9027 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9721 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8224 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8649 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9513 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Schrieffer/Llama-SARM-4B/83b3c073-51a5-4863-a37d-d9ade42a2fc9.json b/data/rewardbench/Schrieffer/Llama-SARM-4B/83b3c073-51a5-4863-a37d-d9ade42a2fc9.json new file mode 100644 index 0000000000000000000000000000000000000000..dae32b4a8228f37b26b02b2fbb2067b5ea6a7426 --- /dev/null +++ b/data/rewardbench/Schrieffer/Llama-SARM-4B/83b3c073-51a5-4863-a37d-d9ade42a2fc9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Schrieffer_Llama-SARM-4B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Schrieffer/Llama-SARM-4B", + "id": "Schrieffer/Llama-SARM-4B", + "developer": "Schrieffer", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7379 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6874 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4281 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9556 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7939 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/e51898d2-b720-42b3-bce5-533624df2e01.json b/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/e51898d2-b720-42b3-bce5-533624df2e01.json new file mode 100644 index 0000000000000000000000000000000000000000..3be0919af7e25264cc94301666cbc1810329316e --- /dev/null +++ b/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/e51898d2-b720-42b3-bce5-533624df2e01.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/ShikaiChen_LDL-Reward-Gemma-2-27B-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1", + "id": "ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1", + "developer": "ShikaiChen", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7249 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7558 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9131 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7633 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/f1ecd2b3-116d-4818-9488-70def015c3e7.json b/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/f1ecd2b3-116d-4818-9488-70def015c3e7.json new file mode 100644 index 0000000000000000000000000000000000000000..91c176635368ec657e5c03f0ceb61f0dfcd00ed5 --- /dev/null +++ b/data/rewardbench/ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1/f1ecd2b3-116d-4818-9488-70def015c3e7.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ShikaiChen_LDL-Reward-Gemma-2-27B-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1", + "id": "ShikaiChen/LDL-Reward-Gemma-2-27B-v0.1", + "developer": "ShikaiChen", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9499 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9079 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9378 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9903 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-70B/181b65b9-921c-43ea-8a16-146514284c8d.json b/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-70B/181b65b9-921c-43ea-8a16-146514284c8d.json new file mode 100644 index 0000000000000000000000000000000000000000..144c486e94965bf7033f7c406d702236f8c33003 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-70B/181b65b9-921c-43ea-8a16-146514284c8d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Critic-Llama-3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Critic-Llama-3.1-70B", + "id": "Skywork/Skywork-Critic-Llama-3.1-70B", + "developer": "Skywork", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9331 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8794 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9311 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9554 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-8B/855dfc51-0daf-4b63-bca4-75ad47c145d1.json b/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-8B/855dfc51-0daf-4b63-bca4-75ad47c145d1.json new file mode 100644 index 0000000000000000000000000000000000000000..ea03c6c8cb6803067ad79f3f823bcf4c3d0eef81 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Critic-Llama-3.1-8B/855dfc51-0daf-4b63-bca4-75ad47c145d1.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Critic-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Critic-Llama-3.1-8B", + "id": "Skywork/Skywork-Critic-Llama-3.1-8B", + "developer": "Skywork", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8896 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9358 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8136 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9108 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.898 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/1a3a8db6-0e82-4ac7-a57e-314967c7abea.json b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/1a3a8db6-0e82-4ac7-a57e-314967c7abea.json new file mode 100644 index 0000000000000000000000000000000000000000..7a5e5f9303cccdfeaa298da596fe42514a0ccf88 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/1a3a8db6-0e82-4ac7-a57e-314967c7abea.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-Gemma-2-27B-v0.2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Gemma-2-27B-v0.2", + "id": "Skywork/Skywork-Reward-Gemma-2-27B-v0.2", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7531 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6721 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9689 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9172 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8182 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/8f1ffd6e-9d15-437d-97aa-734d5b6a3317.json b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/8f1ffd6e-9d15-437d-97aa-734d5b6a3317.json new file mode 100644 index 0000000000000000000000000000000000000000..08492bf492502797256e5e0fcdd58b11857c5068 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B-v0.2/8f1ffd6e-9d15-437d-97aa-734d5b6a3317.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Reward-Gemma-2-27B-v0.2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Gemma-2-27B-v0.2", + "id": "Skywork/Skywork-Reward-Gemma-2-27B-v0.2", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9426 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9609 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8991 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9297 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9807 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/46b36270-a24c-4fc5-b281-612a68104fc5.json b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/46b36270-a24c-4fc5-b281-612a68104fc5.json new file mode 100644 index 0000000000000000000000000000000000000000..57316496c37476518ec689c99fd72003295e7b53 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/46b36270-a24c-4fc5-b281-612a68104fc5.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Reward-Gemma-2-27B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Gemma-2-27B", + "id": "Skywork/Skywork-Reward-Gemma-2-27B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.938 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9145 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9189 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9606 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/a7aab0e4-f27c-4d06-8ebd-c966ab1e50fd.json b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/a7aab0e4-f27c-4d06-8ebd-c966ab1e50fd.json new file mode 100644 index 0000000000000000000000000000000000000000..04dc757e7decb25bd69e55050b14adbebc801e23 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Gemma-2-27B/a7aab0e4-f27c-4d06-8ebd-c966ab1e50fd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-Gemma-2-27B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Gemma-2-27B", + "id": "Skywork/Skywork-Reward-Gemma-2-27B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7576 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7368 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4031 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7049 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9323 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8261 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/abf9b361-f82e-4544-8914-ebb53402f509.json b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/abf9b361-f82e-4544-8914-ebb53402f509.json new file mode 100644 index 0000000000000000000000000000000000000000..58276aa3419d9296444a780767b5fa28776e06cc --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/abf9b361-f82e-4544-8914-ebb53402f509.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-Llama-3.1-8B-v0.2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Llama-3.1-8B-v0.2", + "id": "Skywork/Skywork-Reward-Llama-3.1-8B-v0.2", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7175 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6968 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9414 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7169 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/d4dba5f1-a22e-41ee-a6bb-f565cdb1271d.json b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/d4dba5f1-a22e-41ee-a6bb-f565cdb1271d.json new file mode 100644 index 0000000000000000000000000000000000000000..038a8c89ac0f975e5fbfe158e0be3864064b1ab3 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B-v0.2/d4dba5f1-a22e-41ee-a6bb-f565cdb1271d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Reward-Llama-3.1-8B-v0.2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Llama-3.1-8B-v0.2", + "id": "Skywork/Skywork-Reward-Llama-3.1-8B-v0.2", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9313 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9469 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8838 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.927 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9675 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/c41b5039-7f91-4d5d-95d4-09b1ca03da70.json b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/c41b5039-7f91-4d5d-95d4-09b1ca03da70.json new file mode 100644 index 0000000000000000000000000000000000000000..3a9be487c19450ee538cf8b4c20cee1ac394308b --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/c41b5039-7f91-4d5d-95d4-09b1ca03da70.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-Reward-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Llama-3.1-8B", + "id": "Skywork/Skywork-Reward-Llama-3.1-8B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9252 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8728 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9081 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.962 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/e4b32694-133a-4767-887d-e6898b9282e7.json b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/e4b32694-133a-4767-887d-e6898b9282e7.json new file mode 100644 index 0000000000000000000000000000000000000000..0c1e50b3aaf37834dc5a8ea69c5a81c8257f3463 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-Llama-3.1-8B/e4b32694-133a-4767-887d-e6898b9282e7.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-Llama-3.1-8B", + "id": "Skywork/Skywork-Reward-Llama-3.1-8B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7314 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6989 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.425 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9333 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9616 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.741 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.1-8B/ae354d47-a12c-4978-a094-1f259d986eca.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.1-8B/ae354d47-a12c-4978-a094-1f259d986eca.json new file mode 100644 index 0000000000000000000000000000000000000000..b787f5e2b1204d7841926271272bb68fd18c0c0a --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.1-8B/ae354d47-a12c-4978-a094-1f259d986eca.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Llama-3.1-8B", + "id": "Skywork/Skywork-Reward-V2-Llama-3.1-8B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8413 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8463 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.776 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9838 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8124 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-1B/1ed04bcc-4617-4aea-bb7a-7acccf0b80fa.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-1B/1ed04bcc-4617-4aea-bb7a-7acccf0b80fa.json new file mode 100644 index 0000000000000000000000000000000000000000..6fc52ab22cd74ba28d6297ec6dbd2c229c4c0a9e --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-1B/1ed04bcc-4617-4aea-bb7a-7acccf0b80fa.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Llama-3.2-1B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Llama-3.2-1B", + "id": "Skywork/Skywork-Reward-V2-Llama-3.2-1B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6438 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4562 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8733 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8929 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4306 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-3B/6932e0ef-2302-4072-b8e9-1306c31ee97a.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-3B/6932e0ef-2302-4072-b8e9-1306c31ee97a.json new file mode 100644 index 0000000000000000000000000000000000000000..4b4a2ed53dd7b2c307a063950acb07d59fa296b1 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Llama-3.2-3B/6932e0ef-2302-4072-b8e9-1306c31ee97a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Llama-3.2-3B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Llama-3.2-3B", + "id": "Skywork/Skywork-Reward-V2-Llama-3.2-3B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7466 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7621 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4562 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.694 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9311 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6768 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-0.6B/9b2afcc5-1d4e-41a1-bbb7-ec4e4b908a0e.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-0.6B/9b2afcc5-1d4e-41a1-bbb7-ec4e4b908a0e.json new file mode 100644 index 0000000000000000000000000000000000000000..e5f111fad3163f30d20967664c9677b02acde160 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-0.6B/9b2afcc5-1d4e-41a1-bbb7-ec4e4b908a0e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Qwen3-0.6B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Qwen3-0.6B", + "id": "Skywork/Skywork-Reward-V2-Qwen3-0.6B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6125 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.58 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8444 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7949 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3397 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-1.7B/8c299b3a-14d0-4cc2-a687-346ed1d3ef0d.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-1.7B/8c299b3a-14d0-4cc2-a687-346ed1d3ef0d.json new file mode 100644 index 0000000000000000000000000000000000000000..de7af9b3496dd11c64e9612afe0b162953525adf --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-1.7B/8c299b3a-14d0-4cc2-a687-346ed1d3ef0d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Qwen3-1.7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Qwen3-1.7B", + "id": "Skywork/Skywork-Reward-V2-Qwen3-1.7B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6818 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6568 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7268 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8911 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8848 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4872 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-4B/9cd3db6b-7e57-4a65-b09e-b951a8c82935.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-4B/9cd3db6b-7e57-4a65-b09e-b951a8c82935.json new file mode 100644 index 0000000000000000000000000000000000000000..c486e2d49a63354023eb159e1b37f70f56f6d228 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-4B/9cd3db6b-7e57-4a65-b09e-b951a8c82935.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Qwen3-4B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Qwen3-4B", + "id": "Skywork/Skywork-Reward-V2-Qwen3-4B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7551 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7737 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7322 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9657 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6743 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-8B/0560225f-fd9a-4c67-be67-825a65f8893e.json b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-8B/0560225f-fd9a-4c67-be67-825a65f8893e.json new file mode 100644 index 0000000000000000000000000000000000000000..664c35a44f68642b2d2537f447d3138198f45ad6 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-Reward-V2-Qwen3-8B/0560225f-fd9a-4c67-be67-825a65f8893e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-Reward-V2-Qwen3-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-Reward-V2-Qwen3-8B", + "id": "Skywork/Skywork-Reward-V2-Qwen3-8B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7837 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7989 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7705 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.94 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9636 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7294 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-VL-Reward-7B/43bd329b-13ef-43be-97f6-739764152ca5.json b/data/rewardbench/Skywork/Skywork-VL-Reward-7B/43bd329b-13ef-43be-97f6-739764152ca5.json new file mode 100644 index 0000000000000000000000000000000000000000..7c4b29669fc21579e7bb4c23c5a56c4cff04100d --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-VL-Reward-7B/43bd329b-13ef-43be-97f6-739764152ca5.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/Skywork_Skywork-VL-Reward-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-VL-Reward-7B", + "id": "Skywork/Skywork-VL-Reward-7B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6885 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6063 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8911 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8909 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7586 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/Skywork/Skywork-VL-Reward-7B/44d7fe29-b0bf-4190-b472-c1511d8c3e40.json b/data/rewardbench/Skywork/Skywork-VL-Reward-7B/44d7fe29-b0bf-4190-b472-c1511d8c3e40.json new file mode 100644 index 0000000000000000000000000000000000000000..4b82838acc7e759ccbf1f03a0294ba77cdadc131 --- /dev/null +++ b/data/rewardbench/Skywork/Skywork-VL-Reward-7B/44d7fe29-b0bf-4190-b472-c1511d8c3e40.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/Skywork_Skywork-VL-Reward-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Skywork/Skywork-VL-Reward-7B", + "id": "Skywork/Skywork-VL-Reward-7B", + "developer": "Skywork", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9007 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8994 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.875 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9108 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9176 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/SultanR/SmolTulu-1.7b-RM/d8445594-cc34-4ea9-9d1b-1e99464ca2a8.json b/data/rewardbench/SultanR/SmolTulu-1.7b-RM/d8445594-cc34-4ea9-9d1b-1e99464ca2a8.json new file mode 100644 index 0000000000000000000000000000000000000000..78415ae2cf63cd03d5468c8083641847cb11bcc9 --- /dev/null +++ b/data/rewardbench/SultanR/SmolTulu-1.7b-RM/d8445594-cc34-4ea9-9d1b-1e99464ca2a8.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/SultanR_SmolTulu-1.7b-RM/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "SultanR/SmolTulu-1.7b-RM", + "id": "SultanR/SmolTulu-1.7b-RM", + "developer": "SultanR", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5094 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.743 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4408 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5716 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2821 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ZiyiYe/Con-J-Qwen2-7B/489f57ad-c1dc-43b1-99de-946c5fd21f99.json b/data/rewardbench/ZiyiYe/Con-J-Qwen2-7B/489f57ad-c1dc-43b1-99de-946c5fd21f99.json new file mode 100644 index 0000000000000000000000000000000000000000..6bfbec2307243da7cde0f507e653503e47d2ac8f --- /dev/null +++ b/data/rewardbench/ZiyiYe/Con-J-Qwen2-7B/489f57ad-c1dc-43b1-99de-946c5fd21f99.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ZiyiYe_Con-J-Qwen2-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ZiyiYe/Con-J-Qwen2-7B", + "id": "ZiyiYe/Con-J-Qwen2-7B", + "developer": "ZiyiYe", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8712 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.919 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8026 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8824 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8808 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/llama-2-chat-7b-nectar-3.8m.json/eaad9a99-6354-4128-a707-118eb02beede.json b/data/rewardbench/ai2/llama-2-chat-7b-nectar-3.8m.json/eaad9a99-6354-4128-a707-118eb02beede.json new file mode 100644 index 0000000000000000000000000000000000000000..097f086f18876ce8a76d7d3ca3547761d6c69e57 --- /dev/null +++ b/data/rewardbench/ai2/llama-2-chat-7b-nectar-3.8m.json/eaad9a99-6354-4128-a707-118eb02beede.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_llama-2-chat-7b-nectar-3.8m.json/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/llama-2-chat-7b-nectar-3.8m.json", + "id": "ai2/llama-2-chat-7b-nectar-3.8m.json", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5843 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8631 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2654 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6243 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/llama-2-chat-nectar-180k.json/b43bf7ef-2185-41ed-b92b-10ac6f741a1f.json b/data/rewardbench/ai2/llama-2-chat-nectar-180k.json/b43bf7ef-2185-41ed-b92b-10ac6f741a1f.json new file mode 100644 index 0000000000000000000000000000000000000000..51d9d2f7a247df93c8ca51d79a0e25d752c13fb9 --- /dev/null +++ b/data/rewardbench/ai2/llama-2-chat-nectar-180k.json/b43bf7ef-2185-41ed-b92b-10ac6f741a1f.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_llama-2-chat-nectar-180k.json/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/llama-2-chat-nectar-180k.json", + "id": "ai2/llama-2-chat-nectar-180k.json", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5235 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8827 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2851 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4027 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/llama-2-chat-ultrafeedback-60k.jsonl/2b5520f1-b102-4b70-b5dd-4fb3df8afbc6.json b/data/rewardbench/ai2/llama-2-chat-ultrafeedback-60k.jsonl/2b5520f1-b102-4b70-b5dd-4fb3df8afbc6.json new file mode 100644 index 0000000000000000000000000000000000000000..a03e648956829a9b36a88d939dcb3a5d91e936e6 --- /dev/null +++ b/data/rewardbench/ai2/llama-2-chat-ultrafeedback-60k.jsonl/2b5520f1-b102-4b70-b5dd-4fb3df8afbc6.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_llama-2-chat-ultrafeedback-60k.jsonl/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/llama-2-chat-ultrafeedback-60k.jsonl", + "id": "ai2/llama-2-chat-ultrafeedback-60k.jsonl", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.644 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4539 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5338 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../0db0d5ca-e9f9-460c-8dfb-9de0fe360c4f.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../0db0d5ca-e9f9-460c-8dfb-9de0fe360c4f.json new file mode 100644 index 0000000000000000000000000000000000000000..ba3e2ef145c4a51d5901d96f8970dffb024e203c --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../0db0d5ca-e9f9-460c-8dfb-9de0fe360c4f.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7019 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9497 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7811 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1d032935-71a5-42fa-b091-51fb0d71e8d9.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1d032935-71a5-42fa-b091-51fb0d71e8d9.json new file mode 100644 index 0000000000000000000000000000000000000000..1ef15a4a0b87bc93dd343b0a9f05b1f418705d34 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1d032935-71a5-42fa-b091-51fb0d71e8d9.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6808 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9302 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3596 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7527 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../2f92681d-a630-40fd-862e-c107fea0f359.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../2f92681d-a630-40fd-862e-c107fea0f359.json new file mode 100644 index 0000000000000000000000000000000000000000..249c329bb7b16522aace3096003298681313f8c9 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../2f92681d-a630-40fd-862e-c107fea0f359.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6945 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9385 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3706 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7743 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../3384d497-1ddf-46ea-bb89-5a1751a52fc1.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../3384d497-1ddf-46ea-bb89-5a1751a52fc1.json new file mode 100644 index 0000000000000000000000000000000000000000..175d0f61a06716b5d8085ff53cdf0743c6a3deb9 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../3384d497-1ddf-46ea-bb89-5a1751a52fc1.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6924 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3575 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7757 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5712e669-06a8-4f97-8fa2-de27d0587e9b.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5712e669-06a8-4f97-8fa2-de27d0587e9b.json new file mode 100644 index 0000000000000000000000000000000000000000..8dcd693750e0ede74de6bd151f312c55e7461543 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5712e669-06a8-4f97-8fa2-de27d0587e9b.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7004 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9413 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3882 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7716 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5fc87f26-dee4-4174-9db9-9b614259d8bc.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5fc87f26-dee4-4174-9db9-9b614259d8bc.json new file mode 100644 index 0000000000000000000000000000000000000000..e114ac783dbd06c5c39d3bf4cbedc6b92d23e578 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../5fc87f26-dee4-4174-9db9-9b614259d8bc.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6905 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3596 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7676 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../64b06732-bc3e-40b9-a7ea-164bfcddb1ae.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../64b06732-bc3e-40b9-a7ea-164bfcddb1ae.json new file mode 100644 index 0000000000000000000000000000000000000000..c815d5820b1995a873d29ddb7c3ee0bf9839d0f5 --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../64b06732-bc3e-40b9-a7ea-164bfcddb1ae.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7008 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9385 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3882 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7757 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../6f1dbf5e-c1d9-41c2-8f41-1223da71fe45.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../6f1dbf5e-c1d9-41c2-8f41-1223da71fe45.json new file mode 100644 index 0000000000000000000000000000000000000000..45f8bead0ae0ace510b421ea5761b825e57d62bb --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../6f1dbf5e-c1d9-41c2-8f41-1223da71fe45.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7058 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3947 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7703 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../e2a33114-4a5a-4a23-82e8-6c4c34dc9f77.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../e2a33114-4a5a-4a23-82e8-6c4c34dc9f77.json new file mode 100644 index 0000000000000000000000000000000000000000..15e443b38d5602f522aea9dcc9c4d41e4c7b3c7c --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../e2a33114-4a5a-4a23-82e8-6c4c34dc9f77.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-3.8m-check.../1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-3.8m-check...", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6895 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9385 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3706 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7595 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json/0d58d515-5587-49f4-bf5c-8ef395397228.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json/0d58d515-5587-49f4-bf5c-8ef395397228.json new file mode 100644 index 0000000000000000000000000000000000000000..c71d6fca15f1378365d7d43c288ccab3478dba9f --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json/0d58d515-5587-49f4-bf5c-8ef395397228.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized-700k.json/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7127 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9358 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4079 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7946 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized.json/e343fefb-4939-49bd-bc97-26ba87cdde18.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized.json/e343fefb-4939-49bd-bc97-26ba87cdde18.json new file mode 100644 index 0000000000000000000000000000000000000000..9a603b661edc9b428eb0eb1297ad6ea999e85c2a --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized.json/e343fefb-4939-49bd-bc97-26ba87cdde18.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0-nectar-binarized.json/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0-nectar-binarized.json", + "id": "ai2/tulu-2-7b-rm-v0-nectar-binarized.json", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6756 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9134 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3904 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.723 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/ai2/tulu-2-7b-rm-v0.json/ef92543a-06a6-4714-88fe-234501fd600e.json b/data/rewardbench/ai2/tulu-2-7b-rm-v0.json/ef92543a-06a6-4714-88fe-234501fd600e.json new file mode 100644 index 0000000000000000000000000000000000000000..e747adeda030bc942ba1001308ecad9e01a78ece --- /dev/null +++ b/data/rewardbench/ai2/tulu-2-7b-rm-v0.json/ef92543a-06a6-4714-88fe-234501fd600e.json @@ -0,0 +1,98 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/ai2_tulu-2-7b-rm-v0.json/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "ai2/tulu-2-7b-rm-v0.json", + "id": "ai2/tulu-2-7b-rm-v0.json", + "developer": "ai2", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6655 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.933 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4539 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6095 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/860a3fcd-81a1-4921-b609-9f1522743f7c.json b/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/860a3fcd-81a1-4921-b609-9f1522743f7c.json new file mode 100644 index 0000000000000000000000000000000000000000..d8d61b79836048c00a1b002a96ada027e00e5551 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/860a3fcd-81a1-4921-b609-9f1522743f7c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-70B-Instruct-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-70B-Instruct-RM-RB2", + "id": "allenai/Llama-3.1-70B-Instruct-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7606 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8126 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6995 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8646 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8835 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/c885dd71-2090-46d3-89a5-8302ee2234f4.json b/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/c885dd71-2090-46d3-89a5-8302ee2234f4.json new file mode 100644 index 0000000000000000000000000000000000000000..12aecc90d632a11b8df59fbdfbfead36f74db171 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-70B-Instruct-RM-RB2/c885dd71-2090-46d3-89a5-8302ee2234f4.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-70B-Instruct-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-70B-Instruct-RM-RB2", + "id": "allenai/Llama-3.1-70B-Instruct-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9021 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8355 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9095 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8969 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/105dca01-5fb1-45fc-9965-3d39110968a2.json b/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/105dca01-5fb1-45fc-9965-3d39110968a2.json new file mode 100644 index 0000000000000000000000000000000000000000..3eba55714ea80bb58e31eb044d8f26975675f83f --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/105dca01-5fb1-45fc-9965-3d39110968a2.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-8B-Base-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-8B-Base-RM-RB2", + "id": "allenai/Llama-3.1-8B-Base-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8463 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.933 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7785 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8851 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7886 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/afce506e-e671-4f26-8c1b-d2faff8cfe10.json b/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/afce506e-e671-4f26-8c1b-d2faff8cfe10.json new file mode 100644 index 0000000000000000000000000000000000000000..59ba21f4a5b635e823645c092498c357438875b3 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-8B-Base-RM-RB2/afce506e-e671-4f26-8c1b-d2faff8cfe10.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-8B-Base-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-8B-Base-RM-RB2", + "id": "allenai/Llama-3.1-8B-Base-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.649 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.72 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8267 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8323 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5406 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/8448c145-0fa0-4f1a-9621-250b019fa54d.json b/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/8448c145-0fa0-4f1a-9621-250b019fa54d.json new file mode 100644 index 0000000000000000000000000000000000000000..46324da1904fde89448d3a0e43529c87b8dc2c00 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/8448c145-0fa0-4f1a-9621-250b019fa54d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-8B-Instruct-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-8B-Instruct-RM-RB2", + "id": "allenai/Llama-3.1-8B-Instruct-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7285 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7432 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8956 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9071 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7638 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/fab85638-b889-495b-b7a4-c8729d6e1bfe.json b/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/fab85638-b889-495b-b7a4-c8729d6e1bfe.json new file mode 100644 index 0000000000000000000000000000000000000000..b1ecc8576b5624be02bb24eccb569c63df83e490 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-8B-Instruct-RM-RB2/fab85638-b889-495b-b7a4-c8729d6e1bfe.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-8B-Instruct-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-8B-Instruct-RM-RB2", + "id": "allenai/Llama-3.1-8B-Instruct-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8885 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8932 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.887 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/1b324d99-f032-4c26-98ce-54a2cc0e174c.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/1b324d99-f032-4c26-98ce-54a2cc0e174c.json new file mode 100644 index 0000000000000000000000000000000000000000..1f92d56dceb13d71966c3e50f91e87c7d5f78acd --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/1b324d99-f032-4c26-98ce-54a2cc0e174c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-Tulu-3-70B-SFT-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.722 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6776 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8689 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8308 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/27774b85-1e85-4ee6-b831-359c6ef996a8.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/27774b85-1e85-4ee6-b831-359c6ef996a8.json new file mode 100644 index 0000000000000000000000000000000000000000..e93a100029713cfa42bd2a1d0dba5b59de254ddf --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2/27774b85-1e85-4ee6-b831-359c6ef996a8.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-Tulu-3-70B-SFT-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-70B-SFT-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8892 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8268 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9027 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8583 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/8e8c9694-ade0-4588-9b7c-f63dd1f5a8ee.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/8e8c9694-ade0-4588-9b7c-f63dd1f5a8ee.json new file mode 100644 index 0000000000000000000000000000000000000000..42b817c3442467c174adcae7de91a3284e982888 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/8e8c9694-ade0-4588-9b7c-f63dd1f5a8ee.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-Tulu-3-8B-DPO-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.687 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7516 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.86 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8545 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6397 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/b843867b-817e-423e-a4fd-53dcd5c95c8a.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/b843867b-817e-423e-a4fd-53dcd5c95c8a.json new file mode 100644 index 0000000000000000000000000000000000000000..a91c31fc09b8b7b40c4a6f467c4948e7b750bfa7 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2/b843867b-817e-423e-a4fd-53dcd5c95c8a.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-Tulu-3-8B-DPO-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-DPO-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8431 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.761 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8662 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7898 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/a3e0f4bc-fed1-4066-8da0-142cf2e4f480.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/a3e0f4bc-fed1-4066-8da0-142cf2e4f480.json new file mode 100644 index 0000000000000000000000000000000000000000..9a7ccd0bf65f7f4e1cb4837dd808cfc1646f941d --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/a3e0f4bc-fed1-4066-8da0-142cf2e4f480.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-Tulu-3-8B-RL-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6871 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7642 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8644 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8485 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6281 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/d60f2aaf-c096-42f4-bf33-ac7ee28cd0af.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/d60f2aaf-c096-42f4-bf33-ac7ee28cd0af.json new file mode 100644 index 0000000000000000000000000000000000000000..e904877d10f8e009a169ef0b3ad109f7f3effb42 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2/d60f2aaf-c096-42f4-bf33-ac7ee28cd0af.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-Tulu-3-8B-RL-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-RL-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8369 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9469 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7588 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8703 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7715 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RM/99b1c4bf-dafc-4fb0-92e1-ea2603fff061.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RM/99b1c4bf-dafc-4fb0-92e1-ea2603fff061.json new file mode 100644 index 0000000000000000000000000000000000000000..43a0afbf5ded6a46e841501a54b8ce86eeb9f591 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-RM/99b1c4bf-dafc-4fb0-92e1-ea2603fff061.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-Tulu-3-8B-RM/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-RM", + "id": "allenai/Llama-3.1-Tulu-3-8B-RM", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.59 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7453 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3469 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5364 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5243 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/6efda991-f3f2-4584-8c9c-7b79e3632f00.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/6efda991-f3f2-4584-8c9c-7b79e3632f00.json new file mode 100644 index 0000000000000000000000000000000000000000..840b3907a9a4d7e0e80b6c2a2579b7a8a51e48d8 --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/6efda991-f3f2-4584-8c9c-7b79e3632f00.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_Llama-3.1-Tulu-3-8B-SFT-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8551 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9497 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7917 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8784 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8005 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/ddd0a87e-6a12-4f79-9f4f-e3a8349c90ec.json b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/ddd0a87e-6a12-4f79-9f4f-e3a8349c90ec.json new file mode 100644 index 0000000000000000000000000000000000000000..37be2e58c080be36fc3147531e9498bf41fba49b --- /dev/null +++ b/data/rewardbench/allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2/ddd0a87e-6a12-4f79-9f4f-e3a8349c90ec.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_Llama-3.1-Tulu-3-8B-SFT-RM-RB2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2", + "id": "allenai/Llama-3.1-Tulu-3-8B-SFT-RM-RB2", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6821 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8889 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6063 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/OLMo-7B-Instruct/623f8ad9-efe8-44fb-98e6-b010fd21d985.json b/data/rewardbench/allenai/OLMo-7B-Instruct/623f8ad9-efe8-44fb-98e6-b010fd21d985.json new file mode 100644 index 0000000000000000000000000000000000000000..01f0da939af2c3ad70dec94382cf241d18d1ffa7 --- /dev/null +++ b/data/rewardbench/allenai/OLMo-7B-Instruct/623f8ad9-efe8-44fb-98e6-b010fd21d985.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_OLMo-7B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/OLMo-7B-Instruct", + "id": "allenai/OLMo-7B-Instruct", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6727 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8966 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6486 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7168 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5173 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/llama-3-tulu-2-70b-uf-mean-rm/0c8df4a8-dab9-45f2-ac56-1316db82d71d.json b/data/rewardbench/allenai/llama-3-tulu-2-70b-uf-mean-rm/0c8df4a8-dab9-45f2-ac56-1316db82d71d.json new file mode 100644 index 0000000000000000000000000000000000000000..4577baf2c066baa15bcf8c866ea9d78c45b3448a --- /dev/null +++ b/data/rewardbench/allenai/llama-3-tulu-2-70b-uf-mean-rm/0c8df4a8-dab9-45f2-ac56-1316db82d71d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_llama-3-tulu-2-70b-uf-mean-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/llama-3-tulu-2-70b-uf-mean-rm", + "id": "allenai/llama-3-tulu-2-70b-uf-mean-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7019 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8631 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5614 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6095 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8268 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5957 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/llama-3-tulu-2-8b-uf-mean-rm/316f59aa-4ea4-4c66-8f52-2ac68e4c5013.json b/data/rewardbench/allenai/llama-3-tulu-2-8b-uf-mean-rm/316f59aa-4ea4-4c66-8f52-2ac68e4c5013.json new file mode 100644 index 0000000000000000000000000000000000000000..0970f4dfb8644c5f5589b9cec1cf8410a6627699 --- /dev/null +++ b/data/rewardbench/allenai/llama-3-tulu-2-8b-uf-mean-rm/316f59aa-4ea4-4c66-8f52-2ac68e4c5013.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_llama-3-tulu-2-8b-uf-mean-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/llama-3-tulu-2-8b-uf-mean-rm", + "id": "allenai/llama-3-tulu-2-8b-uf-mean-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7342 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5921 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6162 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8212 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6434 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/llama-3-tulu-2-dpo-70b/c8b6cf8a-b52f-4ba2-b3b7-17ca691f01b8.json b/data/rewardbench/allenai/llama-3-tulu-2-dpo-70b/c8b6cf8a-b52f-4ba2-b3b7-17ca691f01b8.json new file mode 100644 index 0000000000000000000000000000000000000000..e96ea4aee018df2847f53dfe6cd094911b223aa1 --- /dev/null +++ b/data/rewardbench/allenai/llama-3-tulu-2-dpo-70b/c8b6cf8a-b52f-4ba2-b3b7-17ca691f01b8.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_llama-3-tulu-2-dpo-70b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/llama-3-tulu-2-dpo-70b", + "id": "allenai/llama-3-tulu-2-dpo-70b", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7496 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5746 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7486 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.802 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5687 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/llama-3-tulu-2-dpo-8b/16afa67f-bae0-41c8-b592-58575513336e.json b/data/rewardbench/allenai/llama-3-tulu-2-dpo-8b/16afa67f-bae0-41c8-b592-58575513336e.json new file mode 100644 index 0000000000000000000000000000000000000000..25c075aae3227e7872f5207e167963f06c464d74 --- /dev/null +++ b/data/rewardbench/allenai/llama-3-tulu-2-dpo-8b/16afa67f-bae0-41c8-b592-58575513336e.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_llama-3-tulu-2-dpo-8b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/llama-3-tulu-2-dpo-8b", + "id": "allenai/llama-3-tulu-2-dpo-8b", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7275 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5351 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6649 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8663 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5097 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739590997/c13109d7-da42-4260-9b86-386101abaa4a.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739590997/c13109d7-da42-4260-9b86-386101abaa4a.json new file mode 100644 index 0000000000000000000000000000000000000000..847b8934126df59fa64bcdbb1f3b663da8b0465c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739590997/c13109d7-da42-4260-9b86-386101abaa4a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739590997/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739590997", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739590997", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6004 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7032 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7867 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.598 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5165 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739871066/d9912ef2-3d3f-4659-890d-805bf960f6f4.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739871066/d9912ef2-3d3f-4659-890d-805bf960f6f4.json new file mode 100644 index 0000000000000000000000000000000000000000..037804de22e6a02c51c90bd4ffcddf4b51d62b8e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739871066/d9912ef2-3d3f-4659-890d-805bf960f6f4.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739871066/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739871066", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739871066", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6012 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6989 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.425 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.604 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4527 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739925892/ed609bce-7a55-4faa-bca0-f93b5aaf499a.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739925892/ed609bce-7a55-4faa-bca0-f93b5aaf499a.json new file mode 100644 index 0000000000000000000000000000000000000000..ada56a9d0ac0521435fe2238b5d018980a10540d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739925892/ed609bce-7a55-4faa-bca0-f93b5aaf499a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739925892/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739925892", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739925892", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6345 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7432 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8111 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7131 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5606 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943850/9f0b4735-052b-4968-a467-09a5acf78a5d.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943850/9f0b4735-052b-4968-a467-09a5acf78a5d.json new file mode 100644 index 0000000000000000000000000000000000000000..0f4884479ab0eabf422cfb3f6596a37f46487470 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943850/9f0b4735-052b-4968-a467-09a5acf78a5d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739943850/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739943850", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739943850", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4978 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5726 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5191 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6489 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6222 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3114 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943881/63c15bc6-e5a3-4a5d-a22e-d4544d3585e9.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943881/63c15bc6-e5a3-4a5d-a22e-d4544d3585e9.json new file mode 100644 index 0000000000000000000000000000000000000000..e8d9b3beec7695f010cba9d019d6649d2d84375c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943881/63c15bc6-e5a3-4a5d-a22e-d4544d3585e9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739943881/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739943881", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739943881", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5998 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7032 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3187 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6727 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5025 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943972/481b59db-4456-483b-8514-99ae579f8b27.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943972/481b59db-4456-483b-8514-99ae579f8b27.json new file mode 100644 index 0000000000000000000000000000000000000000..69e8a9900f873b459d1b30bf4cc12cde27c908f8 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739943972/481b59db-4456-483b-8514-99ae579f8b27.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739943972/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739943972", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739943972", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5289 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6168 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5738 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5657 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3577 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739957701/9f52f76c-432b-44cc-8790-be0f87e2a71b.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739957701/9f52f76c-432b-44cc-8790-be0f87e2a71b.json new file mode 100644 index 0000000000000000000000000000000000000000..12e1bd28b31b76e2306c0df2ce24662cb58a882a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739957701/9f52f76c-432b-44cc-8790-be0f87e2a71b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739957701/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739957701", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739957701", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6194 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6779 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8022 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.697 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5822 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971507/ee1715d8-4b95-4a17-aea4-7300298ff259.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971507/ee1715d8-4b95-4a17-aea4-7300298ff259.json new file mode 100644 index 0000000000000000000000000000000000000000..5dd02ad94f559138055cc3293e0b77294806b60e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971507/ee1715d8-4b95-4a17-aea4-7300298ff259.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739971507/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739971507", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739971507", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5717 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.68 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5475 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4545 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971529/653ef0be-9765-41ca-b9e9-7d256b7bd13c.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971529/653ef0be-9765-41ca-b9e9-7d256b7bd13c.json new file mode 100644 index 0000000000000000000000000000000000000000..afa66dffbdec882bd9a37837aa4e0ab36ecc3261 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739971529/653ef0be-9765-41ca-b9e9-7d256b7bd13c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739971529/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739971529", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739971529", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5564 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6568 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5956 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7533 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5737 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4027 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739998765/66f6075a-ce94-4f3b-9916-2f5eeac7a16f.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739998765/66f6075a-ce94-4f3b-9916-2f5eeac7a16f.json new file mode 100644 index 0000000000000000000000000000000000000000..447a5d7b6d19cc6217eed2a9027fbc30d74aebfe --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1739998765/66f6075a-ce94-4f3b-9916-2f5eeac7a16f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1739998765/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1739998765", + "id": "allenai/open_instruct_dev-reward_modeling__1__1739998765", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6008 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7095 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8022 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5859 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4883 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740005072/ada5fc02-99ba-4d73-b18a-e173892eb2dc.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740005072/ada5fc02-99ba-4d73-b18a-e173892eb2dc.json new file mode 100644 index 0000000000000000000000000000000000000000..61b795c9469ab5042837c594b9d7ed5ff21645dc --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740005072/ada5fc02-99ba-4d73-b18a-e173892eb2dc.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1740005072/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1740005072", + "id": "allenai/open_instruct_dev-reward_modeling__1__1740005072", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6097 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7137 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6343 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5047 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740129284/5be3bc89-082c-47d2-b006-55e0d3fe6734.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740129284/5be3bc89-082c-47d2-b006-55e0d3fe6734.json new file mode 100644 index 0000000000000000000000000000000000000000..3e8086650471a904b6c8c94bba0a51c63a5ffc73 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1740129284/5be3bc89-082c-47d2-b006-55e0d3fe6734.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1740129284/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1740129284", + "id": "allenai/open_instruct_dev-reward_modeling__1__1740129284", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6129 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7116 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8022 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6101 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4652 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741286813/671b8a9d-6077-4983-82dc-30bb5066b2cb.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741286813/671b8a9d-6077-4983-82dc-30bb5066b2cb.json new file mode 100644 index 0000000000000000000000000000000000000000..34f1ae17d7f6f36cd5ef2954824dd47e8a0fdf76 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741286813/671b8a9d-6077-4983-82dc-30bb5066b2cb.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1741286813/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1741286813", + "id": "allenai/open_instruct_dev-reward_modeling__1__1741286813", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6295 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9111 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8263 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5365 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741287363/fc8deb7b-3fb0-4ca8-8157-899b7ad7a749.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741287363/fc8deb7b-3fb0-4ca8-8157-899b7ad7a749.json new file mode 100644 index 0000000000000000000000000000000000000000..e8563741cdfc2e594a57b52611e6d8807d0ba27e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741287363/fc8deb7b-3fb0-4ca8-8157-899b7ad7a749.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1741287363/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1741287363", + "id": "allenai/open_instruct_dev-reward_modeling__1__1741287363", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6672 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6295 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9374 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5748 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741292911/ba956d3e-152c-46be-ae26-7135b64fd740.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741292911/ba956d3e-152c-46be-ae26-7135b64fd740.json new file mode 100644 index 0000000000000000000000000000000000000000..f457b407fee2095b9d0500ab56f379fa200a7b9a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1741292911/ba956d3e-152c-46be-ae26-7135b64fd740.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1741292911/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1741292911", + "id": "allenai/open_instruct_dev-reward_modeling__1__1741292911", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6607 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6589 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9089 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8869 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5028 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742338142/b9ac9126-38a7-443c-b47b-ea9bf8f67fb0.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742338142/b9ac9126-38a7-443c-b47b-ea9bf8f67fb0.json new file mode 100644 index 0000000000000000000000000000000000000000..ac68d8b5e6a7e1b9ba2886fa25eb5c6c1b28cf9d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742338142/b9ac9126-38a7-443c-b47b-ea9bf8f67fb0.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1742338142/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1742338142", + "id": "allenai/open_instruct_dev-reward_modeling__1__1742338142", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6344 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7049 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.88 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6323 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.475 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519610/f61f3859-74ce-458f-9a65-c58a5ffffb76.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519610/f61f3859-74ce-458f-9a65-c58a5ffffb76.json new file mode 100644 index 0000000000000000000000000000000000000000..1034d44a819b90b77d6a42fa7d19d754938a94e8 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519610/f61f3859-74ce-458f-9a65-c58a5ffffb76.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1742519610/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1742519610", + "id": "allenai/open_instruct_dev-reward_modeling__1__1742519610", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6361 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6721 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.82 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6444 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5915 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519628/4358401c-fddb-4f9f-b245-51932b9dfce2.json b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519628/4358401c-fddb-4f9f-b245-51932b9dfce2.json new file mode 100644 index 0000000000000000000000000000000000000000..ad6d00b384b7dd3712512fe190a920565bfe3189 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-reward_modeling__1__1742519628/4358401c-fddb-4f9f-b245-51932b9dfce2.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-reward_modeling__1__1742519628/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-reward_modeling__1__1742519628", + "id": "allenai/open_instruct_dev-reward_modeling__1__1742519628", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5609 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5179 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8356 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5071 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5254 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455/591d44f1-e40d-4527-a7b7-efc02f8795b4.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455/591d44f1-e40d-4527-a7b7-efc02f8795b4.json new file mode 100644 index 0000000000000000000000000000000000000000..cb28409a187e0cc34c31effcc95d28e79927e0d6 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455/591d44f1-e40d-4527-a7b7-efc02f8795b4.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455", + "id": "allenai/open_instruct_dev-rm_1e-6_1_100pctflipped__1__1744241455", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0576 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.04 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1313 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0546 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0489 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0808 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511/f22a9e2a-1838-45cb-aa54-fb5b0a522c7d.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511/f22a9e2a-1838-45cb-aa54-fb5b0a522c7d.json new file mode 100644 index 0000000000000000000000000000000000000000..04ec4448e4d7084992733a3e2a8e34699cc3e3dd --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511/f22a9e2a-1838-45cb-aa54-fb5b0a522c7d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511", + "id": "allenai/open_instruct_dev-rm_1e-6_1_10pctflipped__1__1743295511", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5499 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6821 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5956 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7356 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5212 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3711 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406/811b5348-f109-479c-833f-e5a6c15a5cd5.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406/811b5348-f109-479c-833f-e5a6c15a5cd5.json new file mode 100644 index 0000000000000000000000000000000000000000..c94d236b0ef6ffa5e268f1a1f022fb3a50635e51 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406/811b5348-f109-479c-833f-e5a6c15a5cd5.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406", + "id": "allenai/open_instruct_dev-rm_1e-6_1_20pctflipped__1__1743295406", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5054 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6358 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6867 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4424 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2922 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136/954fa274-3d0a-47c5-a4e6-169ba10497cd.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136/954fa274-3d0a-47c5-a4e6-169ba10497cd.json new file mode 100644 index 0000000000000000000000000000000000000000..b32aedbef34814fafedf6752f439ce03cfdfcc29 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136/954fa274-3d0a-47c5-a4e6-169ba10497cd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136", + "id": "allenai/open_instruct_dev-rm_1e-6_1_30pctflipped__1__1743325136", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.478 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6442 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6356 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2707 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3496 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398/618b2393-aea4-45c6-a2fc-c09c400f0d4b.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398/618b2393-aea4-45c6-a2fc-c09c400f0d4b.json new file mode 100644 index 0000000000000000000000000000000000000000..21415e28a6f53ed691994d8fc356c081ba2d2073 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398/618b2393-aea4-45c6-a2fc-c09c400f0d4b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398", + "id": "allenai/open_instruct_dev-rm_1e-6_1_50pctflipped__1__1744241398", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.219 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2484 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1717 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.008 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535/c3e9b8b9-21b2-4876-b764-077dc55d9212.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535/c3e9b8b9-21b2-4876-b764-077dc55d9212.json new file mode 100644 index 0000000000000000000000000000000000000000..008c86880e7f15e6f3bfab787ef5edf10035225e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535/c3e9b8b9-21b2-4876-b764-077dc55d9212.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535", + "id": "allenai/open_instruct_dev-rm_1e-6_1_5pctflipped__1__1743444535", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5625 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6821 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7511 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5313 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.403 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo__1__1743550054/d01f1263-f19f-4659-a6c5-bf035f57261a.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo__1__1743550054/d01f1263-f19f-4659-a6c5-bf035f57261a.json new file mode 100644 index 0000000000000000000000000000000000000000..c76664da9720717eeb765815db27a246936bb56b --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo__1__1743550054/d01f1263-f19f-4659-a6c5-bf035f57261a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_dpo__1__1743550054/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_dpo__1__1743550054", + "id": "allenai/open_instruct_dev-rm_1e-6_1_dpo__1__1743550054", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5759 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7578 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5333 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.459 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271/5d97417b-825e-415f-a063-c02d04b32d53.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271/5d97417b-825e-415f-a063-c02d04b32d53.json new file mode 100644 index 0000000000000000000000000000000000000000..781b794a2f95ad630d346c73cc7eb330c1659a9d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271/5d97417b-825e-415f-a063-c02d04b32d53.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271", + "id": "allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworks__1__1744530271", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6057 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5053 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7798 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5419 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181/b4686c6b-ce5b-41d4-bbbf-368189dc8af0.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181/b4686c6b-ce5b-41d4-bbbf-368189dc8af0.json new file mode 100644 index 0000000000000000000000000000000000000000..8533551950c48ea7a45363e30091fd7ea71148ae --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181/b4686c6b-ce5b-41d4-bbbf-368189dc8af0.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181", + "id": "allenai/open_instruct_dev-rm_1e-6_1_dpo_skyworkstulufull__1__1743550181", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6535 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7137 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8244 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7737 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6101 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl__1__1743551221/63057e18-ce38-41e0-b8ed-936ed6b6c5d7.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl__1__1743551221/63057e18-ce38-41e0-b8ed-936ed6b6c5d7.json new file mode 100644 index 0000000000000000000000000000000000000000..9782e0334b37cc977fc72ddc57591dfe64355429 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl__1__1743551221/63057e18-ce38-41e0-b8ed-936ed6b6c5d7.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_rl__1__1743551221/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_rl__1__1743551221", + "id": "allenai/open_instruct_dev-rm_1e-6_1_rl__1__1743551221", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5799 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7116 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.76 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5374 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.461 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262/868b8f2d-e410-421f-bc28-973103590054.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262/868b8f2d-e410-421f-bc28-973103590054.json new file mode 100644 index 0000000000000000000000000000000000000000..bb9399612595b2a75231b5db053ce49ace66125f --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262/868b8f2d-e410-421f-bc28-973103590054.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262", + "id": "allenai/open_instruct_dev-rm_1e-6_1_rl_skyworks__1__1744530262", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5903 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4863 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5738 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8489 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4926 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523/dc9b1bec-b7b9-43c3-8815-524d9b4e9bf2.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523/dc9b1bec-b7b9-43c3-8815-524d9b4e9bf2.json new file mode 100644 index 0000000000000000000000000000000000000000..a24cfa56ddb2c947416d99bb760d94905491f979 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523/dc9b1bec-b7b9-43c3-8815-524d9b4e9bf2.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523", + "id": "allenai/open_instruct_dev-rm_1e-6_1_rl_skyworkstulufull__1__1743551523", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6483 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7758 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6044 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750/5c6b7f8e-d558-4d30-a301-c4e886a930f5.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750/5c6b7f8e-d558-4d30-a301-c4e886a930f5.json new file mode 100644 index 0000000000000000000000000000000000000000..863d31562f6ab8e7f916a96e7a1d176759245ed9 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750/5c6b7f8e-d558-4d30-a301-c4e886a930f5.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750", + "id": "allenai/open_instruct_dev-rm_1e-6_1_skyworkstulumix__1__1743205750", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5157 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7089 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4222 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3791 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427/83ede02b-f1c1-41a7-b456-ffd4bb6d2177.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427/83ede02b-f1c1-41a7-b456-ffd4bb6d2177.json new file mode 100644 index 0000000000000000000000000000000000000000..be7efdb8a78c33702ed8686779adf140be676303 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427/83ede02b-f1c1-41a7-b456-ffd4bb6d2177.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427", + "id": "allenai/open_instruct_dev-rm_1e-6_2_10pctflipped__1__1743295427", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6009 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7273 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3931 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446/80855289-59b7-4983-83d7-3e46be52c12b.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446/80855289-59b7-4983-83d7-3e46be52c12b.json new file mode 100644 index 0000000000000000000000000000000000000000..96a1b9dc411631521159c46b0f010db96a76b36d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446/80855289-59b7-4983-83d7-3e46be52c12b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446", + "id": "allenai/open_instruct_dev-rm_1e-6_2_20pctflipped__1__1743295446", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5716 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6779 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5464 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7533 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3534 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094/6db9f955-42f9-456f-b09b-f95257c8c94b.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094/6db9f955-42f9-456f-b09b-f95257c8c94b.json new file mode 100644 index 0000000000000000000000000000000000000000..f89ddd8b9f3848f59ec2ee9d9e410d459466af72 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094/6db9f955-42f9-456f-b09b-f95257c8c94b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094", + "id": "allenai/open_instruct_dev-rm_1e-6_2_30pctflipped__1__1743325094", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5151 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6484 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5574 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7289 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4889 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3357 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636/9cd80051-4e5c-416b-90dd-899334beb804.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636/9cd80051-4e5c-416b-90dd-899334beb804.json new file mode 100644 index 0000000000000000000000000000000000000000..32497f2e7474fa4e63417673d551382fb8365c1b --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636/9cd80051-4e5c-416b-90dd-899334beb804.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636", + "id": "allenai/open_instruct_dev-rm_1e-6_2_5pctflipped__1__1743444636", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6119 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.72 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8067 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6889 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.421 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_dpo__1__1743549325/6bef2e1a-a7be-4ae7-9dfe-82079bd92ff6.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_dpo__1__1743549325/6bef2e1a-a7be-4ae7-9dfe-82079bd92ff6.json new file mode 100644 index 0000000000000000000000000000000000000000..01e1ee0a18746b93441b12663a45c5095abc59b0 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_dpo__1__1743549325/6bef2e1a-a7be-4ae7-9dfe-82079bd92ff6.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_dpo__1__1743549325/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_dpo__1__1743549325", + "id": "allenai/open_instruct_dev-rm_1e-6_2_dpo__1__1743549325", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6008 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7179 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5956 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6707 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4707 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_rl__1__1743551238/3a274ba9-8858-41b9-ae77-93c6ef5ad170.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_rl__1__1743551238/3a274ba9-8858-41b9-ae77-93c6ef5ad170.json new file mode 100644 index 0000000000000000000000000000000000000000..8d41ab99f5828f0021a37a0517edccd3cf12307d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_rl__1__1743551238/3a274ba9-8858-41b9-ae77-93c6ef5ad170.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_rl__1__1743551238/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_rl__1__1743551238", + "id": "allenai/open_instruct_dev-rm_1e-6_2_rl__1__1743551238", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5965 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7095 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8044 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6566 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.453 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906/abc33cc5-184d-42eb-9abb-ce35d4f137be.json b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906/abc33cc5-184d-42eb-9abb-ce35d4f137be.json new file mode 100644 index 0000000000000000000000000000000000000000..8da1c83a1a440c54044475a14846d028772b7a82 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906/abc33cc5-184d-42eb-9abb-ce35d4f137be.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906", + "id": "allenai/open_instruct_dev-rm_1e-6_2_skyworkstulumix__1__1743205906", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5574 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6526 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7711 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4208 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529/59421cf3-4025-4891-8a0c-596b2850db52.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529/59421cf3-4025-4891-8a0c-596b2850db52.json new file mode 100644 index 0000000000000000000000000000000000000000..a9e4a695c267d074fe1b687815ddd5e9f789526c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529/59421cf3-4025-4891-8a0c-596b2850db52.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529", + "id": "allenai/open_instruct_dev-rm_2e-5_1_100pctflipped__1__1744241529", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0719 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0421 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0601 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0378 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0949 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305/063c864c-d38e-4756-a306-24c814713235.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305/063c864c-d38e-4756-a306-24c814713235.json new file mode 100644 index 0000000000000000000000000000000000000000..2f290f678cbed6d0c894d62f010a0d0fa14ea900 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305/063c864c-d38e-4756-a306-24c814713235.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305", + "id": "allenai/open_instruct_dev-rm_2e-5_1_10pctflipped__1__1743295305", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.553 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6733 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5697 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4227 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778/f4fb519b-3674-4df2-a050-672d0694be32.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778/f4fb519b-3674-4df2-a050-672d0694be32.json new file mode 100644 index 0000000000000000000000000000000000000000..d1813426435af9f8dc2610e8e3624a7d9a0cdc51 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778/f4fb519b-3674-4df2-a050-672d0694be32.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778", + "id": "allenai/open_instruct_dev-rm_2e-5_1_20pctflipped__1__1743324778", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4955 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6189 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.325 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6378 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5657 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2466 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459/ec251610-c326-46e2-81e9-5200c9413cc6.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459/ec251610-c326-46e2-81e9-5200c9413cc6.json new file mode 100644 index 0000000000000000000000000000000000000000..a0ead88c1c2a3a60aae143ce5529f627f4abd0ec --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459/ec251610-c326-46e2-81e9-5200c9413cc6.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459", + "id": "allenai/open_instruct_dev-rm_2e-5_1_30pctflipped__1__1743326459", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4198 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5747 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5464 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2073 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747/cb30c7a2-3f94-4f5d-91f6-8be5a4d87855.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747/cb30c7a2-3f94-4f5d-91f6-8be5a4d87855.json new file mode 100644 index 0000000000000000000000000000000000000000..f93ef46a4c3e54317499afaec6e28ac5c5d769f4 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747/cb30c7a2-3f94-4f5d-91f6-8be5a4d87855.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747", + "id": "allenai/open_instruct_dev-rm_2e-5_1_5pctflipped__1__1743443747", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5465 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6821 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7333 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3713 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935/6e825bc2-e022-4feb-ae16-b2d31833906b.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935/6e825bc2-e022-4feb-ae16-b2d31833906b.json new file mode 100644 index 0000000000000000000000000000000000000000..fbbc5f3cb907419b1ad8bad379300d43e327ddfa --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935/6e825bc2-e022-4feb-ae16-b2d31833906b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935", + "id": "allenai/open_instruct_dev-rm_2e-5_1_skyworkstulumix__1__1743205935", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5197 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6126 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5847 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7333 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4646 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3855 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360/b567ca04-cc5d-406b-bce4-fd6bcfdbd70b.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360/b567ca04-cc5d-406b-bce4-fd6bcfdbd70b.json new file mode 100644 index 0000000000000000000000000000000000000000..f0ded35c13d90108c245c6519a1c18e5db231298 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360/b567ca04-cc5d-406b-bce4-fd6bcfdbd70b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360", + "id": "allenai/open_instruct_dev-rm_2e-5_2_10pctflipped__1__1743295360", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4555 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5495 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3063 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4262 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5711 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6101 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2696 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366/5176d625-7247-4145-ad60-81e5c8be7cbb.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366/5176d625-7247-4145-ad60-81e5c8be7cbb.json new file mode 100644 index 0000000000000000000000000000000000000000..14f6d09b9688bb66afbeac1e2c8f112a38b304e8 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366/5176d625-7247-4145-ad60-81e5c8be7cbb.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366", + "id": "allenai/open_instruct_dev-rm_2e-5_2_20pctflipped__1__1743295366", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4422 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5053 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4044 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6646 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1991 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352/a62210e7-2e97-4f36-9316-e365f6d4982d.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352/a62210e7-2e97-4f36-9316-e365f6d4982d.json new file mode 100644 index 0000000000000000000000000000000000000000..922142065d5e61d5f8bad03e5a9fa5ec0ebf03e1 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352/a62210e7-2e97-4f36-9316-e365f6d4982d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352", + "id": "allenai/open_instruct_dev-rm_2e-5_2_30pctflipped__1__1743326352", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.341 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3333 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3711 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3919 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.195 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634/1802b486-32f2-4f6f-931f-3fcf4a8a83c3.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634/1802b486-32f2-4f6f-931f-3fcf4a8a83c3.json new file mode 100644 index 0000000000000000000000000000000000000000..394b8a5b459b4fc561c4ee08ac9f935eccfdd828 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634/1802b486-32f2-4f6f-931f-3fcf4a8a83c3.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634", + "id": "allenai/open_instruct_dev-rm_2e-5_2_5pctflipped__1__1743444634", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4698 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5853 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2562 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5027 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6489 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5697 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2562 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988/cf5ce4cc-f2c2-43b3-9aa0-72942b5bf759.json b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988/cf5ce4cc-f2c2-43b3-9aa0-72942b5bf759.json new file mode 100644 index 0000000000000000000000000000000000000000..a0a2351796183416bed72eea1f6c088a0e422ee6 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988/cf5ce4cc-f2c2-43b3-9aa0-72942b5bf759.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988", + "id": "allenai/open_instruct_dev-rm_2e-5_2_skyworkstulumix__1__1743205988", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4791 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6421 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.541 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6911 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4182 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.27 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103/36fb9c89-e904-40c5-b00c-935f5d60e45f.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103/36fb9c89-e904-40c5-b00c-935f5d60e45f.json new file mode 100644 index 0000000000000000000000000000000000000000..7bc0e89f998770b88fcb4a68af2afee58b44b123 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103/36fb9c89-e904-40c5-b00c-935f5d60e45f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103", + "id": "allenai/open_instruct_dev-rm_3e-6_1_100pctflipped__1__1744242103", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0607 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0274 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0656 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.04 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0788 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": -0.01 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835/92cfd989-7ec9-43a0-82b1-35e66c34b2d9.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835/92cfd989-7ec9-43a0-82b1-35e66c34b2d9.json new file mode 100644 index 0000000000000000000000000000000000000000..87cb6879d81ef20b1b184bfe1bd2ee4ed89cd2d9 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835/92cfd989-7ec9-43a0-82b1-35e66c34b2d9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835", + "id": "allenai/open_instruct_dev-rm_3e-6_1_10pctflipped__1__1743324835", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6089 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7284 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7622 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6444 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4686 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221/91dd6678-14cc-420a-9cd5-c7e714af018a.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221/91dd6678-14cc-420a-9cd5-c7e714af018a.json new file mode 100644 index 0000000000000000000000000000000000000000..193e2c16280e81d2b8ba1c93fc5cf7c3b4b49453 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221/91dd6678-14cc-420a-9cd5-c7e714af018a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221", + "id": "allenai/open_instruct_dev-rm_3e-6_1_1pctflipped__1__1743445221", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6032 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5859 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5051 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826/ab6541da-98f2-49bb-8abd-ffe7012423fc.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826/ab6541da-98f2-49bb-8abd-ffe7012423fc.json new file mode 100644 index 0000000000000000000000000000000000000000..2007f8f54f1f76a9369fc4d050a71a74b9230dc3 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826/ab6541da-98f2-49bb-8abd-ffe7012423fc.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826", + "id": "allenai/open_instruct_dev-rm_3e-6_1_20pctflipped__1__1743324826", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5831 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6947 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.74 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5758 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4465 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363/333cdc08-4e6d-4aa5-8c4e-d744e0ee841c.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363/333cdc08-4e6d-4aa5-8c4e-d744e0ee841c.json new file mode 100644 index 0000000000000000000000000000000000000000..2e3028502fc6bc1ea92805d344b2502a35a8deb3 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363/333cdc08-4e6d-4aa5-8c4e-d744e0ee841c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363", + "id": "allenai/open_instruct_dev-rm_3e-6_1_30pctflipped__1__1743326363", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5268 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.68 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4343 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3809 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498/9f0c7b72-3077-4837-a53e-b3039a228996.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498/9f0c7b72-3077-4837-a53e-b3039a228996.json new file mode 100644 index 0000000000000000000000000000000000000000..8736b1f1352cd8b6ab0548bdf4f200f0344b7593 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498/9f0c7b72-3077-4837-a53e-b3039a228996.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498", + "id": "allenai/open_instruct_dev-rm_3e-6_1_5pctflipped__1__1743444498", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6093 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4313 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7578 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5859 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5143 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__2__1743897475/ea9fb7fc-f88f-4c8b-b346-b8cbcaad5b09.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__2__1743897475/ea9fb7fc-f88f-4c8b-b346-b8cbcaad5b09.json new file mode 100644 index 0000000000000000000000000000000000000000..bb6d8a281432afd220974d7d5f7cc604093997a0 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__2__1743897475/ea9fb7fc-f88f-4c8b-b346-b8cbcaad5b09.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1__2__1743897475/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1__2__1743897475", + "id": "allenai/open_instruct_dev-rm_3e-6_1__2__1743897475", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6122 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7368 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8044 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.602 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5071 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__3__1744311421/1fb27d96-ce07-464c-8386-44fdab7d396d.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__3__1744311421/1fb27d96-ce07-464c-8386-44fdab7d396d.json new file mode 100644 index 0000000000000000000000000000000000000000..d77a7afdc5dc31fd48713c10a8acaa8539ec567d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1__3__1744311421/1fb27d96-ce07-464c-8386-44fdab7d396d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1__3__1744311421/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1__3__1744311421", + "id": "allenai/open_instruct_dev-rm_3e-6_1__3__1744311421", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5995 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7179 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6323 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.503 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo__1__1743549903/bbcee623-92e4-41ab-9d13-6eaca001c77c.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo__1__1743549903/bbcee623-92e4-41ab-9d13-6eaca001c77c.json new file mode 100644 index 0000000000000000000000000000000000000000..4b0bf8baba4220f467e2c6d0469bd39734ca5b88 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo__1__1743549903/bbcee623-92e4-41ab-9d13-6eaca001c77c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_dpo__1__1743549903/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_dpo__1__1743549903", + "id": "allenai/open_instruct_dev-rm_3e-6_1_dpo__1__1743549903", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6154 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6061 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5043 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368/2c7dc54f-2126-420b-a78f-52232a3ac1e6.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368/2c7dc54f-2126-420b-a78f-52232a3ac1e6.json new file mode 100644 index 0000000000000000000000000000000000000000..20435d15fc2526964ef509dadedbff6a766c09e2 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368/2c7dc54f-2126-420b-a78f-52232a3ac1e6.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368", + "id": "allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworks__1__1744530368", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6604 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6316 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9044 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8929 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5604 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182/a6e652b0-2584-4415-99aa-26ffa4dbd7f8.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182/a6e652b0-2584-4415-99aa-26ffa4dbd7f8.json new file mode 100644 index 0000000000000000000000000000000000000000..a31854d93fb92d03fdf4aa73b63e6977761d6dd1 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182/a6e652b0-2584-4415-99aa-26ffa4dbd7f8.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182", + "id": "allenai/open_instruct_dev-rm_3e-6_1_dpo_skyworkstulufull__1__1743550182", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6783 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7705 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.84 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8101 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6427 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__2__1744316012/7ee2af40-fab5-4b82-8eae-cbebcb46a87a.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__2__1744316012/7ee2af40-fab5-4b82-8eae-cbebcb46a87a.json new file mode 100644 index 0000000000000000000000000000000000000000..9a581c97e3bc131eac559b62f645b43e26917d15 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__2__1744316012/7ee2af40-fab5-4b82-8eae-cbebcb46a87a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_no_if__2__1744316012/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_no_if__2__1744316012", + "id": "allenai/open_instruct_dev-rm_3e-6_1_no_if__2__1744316012", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5911 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7347 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.74 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.604 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4392 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__3__1744315765/e03af491-c5fc-405e-be17-9daad7ae71f4.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__3__1744315765/e03af491-c5fc-405e-be17-9daad7ae71f4.json new file mode 100644 index 0000000000000000000000000000000000000000..d24b4fc9888fd3879e7f813b48a898635397be3e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_no_if__3__1744315765/e03af491-c5fc-405e-be17-9daad7ae71f4.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_no_if__3__1744315765/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_no_if__3__1744315765", + "id": "allenai/open_instruct_dev-rm_3e-6_1_no_if__3__1744315765", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5926 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7889 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5879 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4733 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl__1__1743551527/2aa4a0bd-d8f7-42e7-b081-1047913fe7fc.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl__1__1743551527/2aa4a0bd-d8f7-42e7-b081-1047913fe7fc.json new file mode 100644 index 0000000000000000000000000000000000000000..47e0f9b90a42c9456f3f5552870bc093a1abf388 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl__1__1743551527/2aa4a0bd-d8f7-42e7-b081-1047913fe7fc.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_rl__1__1743551527/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_rl__1__1743551527", + "id": "allenai/open_instruct_dev-rm_3e-6_1_rl__1__1743551527", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6126 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7411 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.425 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7822 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5939 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5104 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236/fb10367a-56cf-4543-9d6d-6c408c50b7e7.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236/fb10367a-56cf-4543-9d6d-6c408c50b7e7.json new file mode 100644 index 0000000000000000000000000000000000000000..d7a9eaa0df8282813d7edf12e83970218db966a5 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236/fb10367a-56cf-4543-9d6d-6c408c50b7e7.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236", + "id": "allenai/open_instruct_dev-rm_3e-6_1_rl_skyworks__1__1744530236", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6525 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6021 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8626 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.59 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530/64933396-ef2c-441e-a5b5-9df21c3ec39d.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530/64933396-ef2c-441e-a5b5-9df21c3ec39d.json new file mode 100644 index 0000000000000000000000000000000000000000..0d42c5f4e38572f9c48b17d395f175a2374f9938 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530/64933396-ef2c-441e-a5b5-9df21c3ec39d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530", + "id": "allenai/open_instruct_dev-rm_3e-6_1_rl_skyworkstulufull__1__1743551530", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6849 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7453 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8404 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6885 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417/5475b38f-f43c-4b5e-8d08-9168c149155f.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417/5475b38f-f43c-4b5e-8d08-9168c149155f.json new file mode 100644 index 0000000000000000000000000000000000000000..44538bbce98529b224f78107eef81b5a414fa0d7 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417/5475b38f-f43c-4b5e-8d08-9168c149155f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417", + "id": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulu75__1__1743534417", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.586 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6632 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.425 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5172 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.477 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486/e091a88c-47a4-4891-83b8-a0377f8c4937.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486/e091a88c-47a4-4891-83b8-a0377f8c4937.json new file mode 100644 index 0000000000000000000000000000000000000000..2499b71bd92c641afc1ea3db9636b2d1ebacf6af --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486/e091a88c-47a4-4891-83b8-a0377f8c4937.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486", + "id": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__1__1743446486", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6773 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7432 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.804 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6626 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745/e94c89cf-3c8d-405e-bf70-bedf73ff3bfd.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745/e94c89cf-3c8d-405e-bf70-bedf73ff3bfd.json new file mode 100644 index 0000000000000000000000000000000000000000..b2f1086216dd4a5572aae4868c0b6618218277f6 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745/e94c89cf-3c8d-405e-bf70-bedf73ff3bfd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745", + "id": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__2__1744314745", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6793 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7558 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8311 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8061 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6485 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661/ff1f577c-a691-45fe-8c27-37b0990ea05b.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661/ff1f577c-a691-45fe-8c27-37b0990ea05b.json new file mode 100644 index 0000000000000000000000000000000000000000..976e953dd46d8f01547fa7be4dff8385926904d4 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661/ff1f577c-a691-45fe-8c27-37b0990ea05b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661", + "id": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulufull__3__1744311661", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6611 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.72 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6393 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8444 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7636 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6428 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472/b5cc06ed-357d-4635-a6d8-b33c90f42ee3.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472/b5cc06ed-357d-4635-a6d8-b33c90f42ee3.json new file mode 100644 index 0000000000000000000000000000000000000000..c799818c5af85bc58cc75dc69215c4e2dbdf3d30 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472/b5cc06ed-357d-4635-a6d8-b33c90f42ee3.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472", + "id": "allenai/open_instruct_dev-rm_3e-6_1_skyworkstulumix__1__1743204472", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5778 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5172 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5003 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267/73c18caa-3e55-44c1-9d0f-f91e52e37dfe.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267/73c18caa-3e55-44c1-9d0f-f91e52e37dfe.json new file mode 100644 index 0000000000000000000000000000000000000000..cf701eb2ac1b930ea0e17a45a13e51bc114f6eae --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267/73c18caa-3e55-44c1-9d0f-f91e52e37dfe.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267", + "id": "allenai/open_instruct_dev-rm_3e-6_2_10pctflipped__1__1743295267", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5746 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6505 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5082 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7414 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4128 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759/9504aaf3-1fd7-4d1e-ad26-ba2c2292ce53.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759/9504aaf3-1fd7-4d1e-ad26-ba2c2292ce53.json new file mode 100644 index 0000000000000000000000000000000000000000..12d4d53684ee10f153d1bb9715a993d66b53026e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759/9504aaf3-1fd7-4d1e-ad26-ba2c2292ce53.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759", + "id": "allenai/open_instruct_dev-rm_3e-6_2_1pctflipped__1__1743445759", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6065 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7116 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5792 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7152 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.465 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905/d79a6349-e46e-44c5-a859-3c817e1c69af.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905/d79a6349-e46e-44c5-a859-3c817e1c69af.json new file mode 100644 index 0000000000000000000000000000000000000000..f229d4011ca15bdd97f52a5b22643da6f3f1bc5d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905/d79a6349-e46e-44c5-a859-3c817e1c69af.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905", + "id": "allenai/open_instruct_dev-rm_3e-6_2_20pctflipped__1__1743324905", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5305 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5832 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.459 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7071 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3849 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363/65962fc8-577f-4ed1-a561-0ad3c569c6fa.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363/65962fc8-577f-4ed1-a561-0ad3c569c6fa.json new file mode 100644 index 0000000000000000000000000000000000000000..471e590bc87df543eea06c60fee744bf34673a86 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363/65962fc8-577f-4ed1-a561-0ad3c569c6fa.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363", + "id": "allenai/open_instruct_dev-rm_3e-6_2_30pctflipped__1__1743326363", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4436 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5411 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3115 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6267 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5414 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.31 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505/9af375ae-939a-4ee6-959b-20e06c198de0.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505/9af375ae-939a-4ee6-959b-20e06c198de0.json new file mode 100644 index 0000000000000000000000000000000000000000..811f5abcdf46ade8dc5bee6aba57ed6e330a383d --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505/9af375ae-939a-4ee6-959b-20e06c198de0.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505", + "id": "allenai/open_instruct_dev-rm_3e-6_2_5pctflipped__1__1743444505", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5925 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.68 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5519 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.78 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7434 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.431 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo__1__1743550180/4c13f3cd-4533-4e30-b12b-68f07be2b53f.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo__1__1743550180/4c13f3cd-4533-4e30-b12b-68f07be2b53f.json new file mode 100644 index 0000000000000000000000000000000000000000..19ed39d14faa2eb65d3600f8c1d115e139a55eef --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo__1__1743550180/4c13f3cd-4533-4e30-b12b-68f07be2b53f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_dpo__1__1743550180/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_dpo__1__1743550180", + "id": "allenai/open_instruct_dev-rm_3e-6_2_dpo__1__1743550180", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6198 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8133 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7232 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4908 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187/32ddead2-292d-4d0d-b52a-0b84ae351bb1.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187/32ddead2-292d-4d0d-b52a-0b84ae351bb1.json new file mode 100644 index 0000000000000000000000000000000000000000..bebe592985cfe4365117159f132885860d287e7a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187/32ddead2-292d-4d0d-b52a-0b84ae351bb1.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187", + "id": "allenai/open_instruct_dev-rm_3e-6_2_dpo_skyworkstulufull__1__1743550187", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6763 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7411 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8545 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5908 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl__1__1743551509/3ecc5ee8-6d06-42bf-bbd7-ad483bc9286f.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl__1__1743551509/3ecc5ee8-6d06-42bf-bbd7-ad483bc9286f.json new file mode 100644 index 0000000000000000000000000000000000000000..9c7e42c3a3ceaa047cb226fcc6f6e923e7471805 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl__1__1743551509/3ecc5ee8-6d06-42bf-bbd7-ad483bc9286f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_rl__1__1743551509/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_rl__1__1743551509", + "id": "allenai/open_instruct_dev-rm_3e-6_2_rl__1__1743551509", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6245 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7242 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8178 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7253 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5124 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498/295fa653-7361-4105-8ef9-89f48c878fdf.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498/295fa653-7361-4105-8ef9-89f48c878fdf.json new file mode 100644 index 0000000000000000000000000000000000000000..05b2670ce1353993b15073c19f0c52e5b33a3c9c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498/295fa653-7361-4105-8ef9-89f48c878fdf.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498", + "id": "allenai/open_instruct_dev-rm_3e-6_2_rl_skyworkstulufull__1__1743551498", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6673 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8622 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8566 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5911 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926/3a69ed02-d4d3-4297-9cbd-614ab2cf9296.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926/3a69ed02-d4d3-4297-9cbd-614ab2cf9296.json new file mode 100644 index 0000000000000000000000000000000000000000..caef35b4cfcf30aec9a7aaec7d43d26a7f93b709 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926/3a69ed02-d4d3-4297-9cbd-614ab2cf9296.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926", + "id": "allenai/open_instruct_dev-rm_3e-6_2_skyworkstulu75__1__1743548926", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5863 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5515 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4768 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661/52e8583a-f3d5-4745-b0d5-f2667101c73b.json b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661/52e8583a-f3d5-4745-b0d5-f2667101c73b.json new file mode 100644 index 0000000000000000000000000000000000000000..36aa03239103e4475ac61cf6ac19f22fc1b51e0c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661/52e8583a-f3d5-4745-b0d5-f2667101c73b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661", + "id": "allenai/open_instruct_dev-rm_3e-6_2_skyworkstulumix__1__1743205661", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.589 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6842 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6393 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7867 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6081 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.447 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598/d2c9ad5f-f187-42d0-974b-d14222c2b7ad.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598/d2c9ad5f-f187-42d0-974b-d14222c2b7ad.json new file mode 100644 index 0000000000000000000000000000000000000000..b49d07351dc0f05f2b860dfbdd15a5829ef0e902 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598/d2c9ad5f-f187-42d0-974b-d14222c2b7ad.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598", + "id": "allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__4__1747266598", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7306 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7474 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.694 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8622 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8061 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8992 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923/f96bc10f-6456-4136-8af3-5025a41d6161.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923/f96bc10f-6456-4136-8af3-5025a41d6161.json new file mode 100644 index 0000000000000000000000000000000000000000..19f9a75456ffdecb3649dca62d3f3db20fec443a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923/f96bc10f-6456-4136-8af3-5025a41d6161.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923", + "id": "allenai/open_instruct_dev-rm_llama70b_skyworkstulufull__8__1745387923", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7573 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8168 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7049 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8733 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8545 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8814 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1__1__1743896628/f86be0c2-ee28-424d-af53-9a81346f4905.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1__1__1743896628/f86be0c2-ee28-424d-af53-9a81346f4905.json new file mode 100644 index 0000000000000000000000000000000000000000..72be57d5c0da183c1b98c2e5fa22df4cab6db47a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1__1__1743896628/f86be0c2-ee28-424d-af53-9a81346f4905.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_1e-6_1__1__1743896628/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_1e-6_1__1__1743896628", + "id": "allenai/open_instruct_dev-rm_llama_1e-6_1__1__1743896628", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6637 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6947 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7273 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6834 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999/4150f740-e67a-4eaa-8213-573e9816c1a2.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999/4150f740-e67a-4eaa-8213-573e9816c1a2.json new file mode 100644 index 0000000000000000000000000000000000000000..8386f384c30b520d436bc468292ac7870a3a0939 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999/4150f740-e67a-4eaa-8213-573e9816c1a2.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999", + "id": "allenai/open_instruct_dev-rm_llama_1e-6_1_skyworks__1__1744062999", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6665 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5979 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8956 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8606 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6422 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777/56e7d1c2-dbd6-4448-a8f8-79e705527c72.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777/56e7d1c2-dbd6-4448-a8f8-79e705527c72.json new file mode 100644 index 0000000000000000000000000000000000000000..f59765b092e3f72309e744b3147f627bd43a8ab9 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777/56e7d1c2-dbd6-4448-a8f8-79e705527c72.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777", + "id": "allenai/open_instruct_dev-rm_llama_1e-6_1_skyworkstulufull__1__1743712777", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7038 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6947 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8867 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8586 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7331 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2__1__1743896638/a10dbff7-075a-4540-94a1-24a455102ec2.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2__1__1743896638/a10dbff7-075a-4540-94a1-24a455102ec2.json new file mode 100644 index 0000000000000000000000000000000000000000..f57480481a0ff9d8eeabbabab73c9002e4b798bf --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2__1__1743896638/a10dbff7-075a-4540-94a1-24a455102ec2.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_1e-6_2__1__1743896638/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_1e-6_2__1__1743896638", + "id": "allenai/open_instruct_dev-rm_llama_1e-6_2__1__1743896638", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6754 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6716 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8756 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7737 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6976 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938/a65655ca-d8c5-418b-bdc3-d31cfcc76e3e.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938/a65655ca-d8c5-418b-bdc3-d31cfcc76e3e.json new file mode 100644 index 0000000000000000000000000000000000000000..b93ff9df11437c6cd6a1eb6566d7f5074ed5b28f --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938/a65655ca-d8c5-418b-bdc3-d31cfcc76e3e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938", + "id": "allenai/open_instruct_dev-rm_llama_1e-6_2_skyworkstulufull__1__1743800938", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7241 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7305 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6667 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9414 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6635 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885/d2a9559c-f400-4429-a7a8-2bca1b446b05.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885/d2a9559c-f400-4429-a7a8-2bca1b446b05.json new file mode 100644 index 0000000000000000000000000000000000000000..5165d8b48e86bf9213ebb2dbe70d970d45312342 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885/d2a9559c-f400-4429-a7a8-2bca1b446b05.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885", + "id": "allenai/open_instruct_dev-rm_llama_2e-5_1_skyworkstulufull__1__1743712885", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6716 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6632 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.82 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8303 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.719 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773/d3cfe789-8bd4-4638-a486-d108fa2c2a10.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773/d3cfe789-8bd4-4638-a486-d108fa2c2a10.json new file mode 100644 index 0000000000000000000000000000000000000000..834cf2f986b3e5e1af64534d145bd92e588f2557 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773/d3cfe789-8bd4-4638-a486-d108fa2c2a10.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773", + "id": "allenai/open_instruct_dev-rm_llama_2e-5_2_skyworkstulufull__1__1743800773", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6207 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6358 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.375 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8267 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.802 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4948 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867/4bc24c36-f62d-4499-952b-27dfb35cc675.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867/4bc24c36-f62d-4499-952b-27dfb35cc675.json new file mode 100644 index 0000000000000000000000000000000000000000..60b7004931e96c2fb07d182849b5a2f1a10ba69f --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867/4bc24c36-f62d-4499-952b-27dfb35cc675.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867", + "id": "allenai/open_instruct_dev-rm_llama_2e-6_1_skyworkstulufull__1__1743893867", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.719 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6393 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8956 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9273 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.738 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__1__1743929424/96171df8-6f06-4e4c-87eb-8449965ca4fa.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__1__1743929424/96171df8-6f06-4e4c-87eb-8449965ca4fa.json new file mode 100644 index 0000000000000000000000000000000000000000..2e5f77eb9c12f601f59d71d90f2b04cfc7518e35 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__1__1743929424/96171df8-6f06-4e4c-87eb-8449965ca4fa.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1__1__1743929424/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1__1__1743929424", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1__1__1743929424", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6572 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7305 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8289 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.703 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6837 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__2__1744311395/18081f23-6348-4d74-835f-4b72b7228afe.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__2__1744311395/18081f23-6348-4d74-835f-4b72b7228afe.json new file mode 100644 index 0000000000000000000000000000000000000000..5e59897cea5ffa32afd579e1347279142d73b56f --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__2__1744311395/18081f23-6348-4d74-835f-4b72b7228afe.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1__2__1744311395/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1__2__1744311395", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1__2__1744311395", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6938 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7537 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.45 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6393 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7616 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6913 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__3__1744311491/ca77b32c-11dc-4303-9229-ca31070e2249.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__3__1744311491/ca77b32c-11dc-4303-9229-ca31070e2249.json new file mode 100644 index 0000000000000000000000000000000000000000..3c8298d9731790f13e88be30362fae1cb4419462 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1__3__1744311491/ca77b32c-11dc-4303-9229-ca31070e2249.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1__3__1744311491/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1__3__1744311491", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1__3__1744311491", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6754 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7242 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7535 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6976 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787/bea46a0e-7e8e-40a1-a782-41b12e8529af.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787/bea46a0e-7e8e-40a1-a782-41b12e8529af.json new file mode 100644 index 0000000000000000000000000000000000000000..ea9edebf1bc7c667ea10e311c8a43284737cd5ab --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787/bea46a0e-7e8e-40a1-a782-41b12e8529af.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworks__1__1744062787", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7045 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6253 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6667 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.92 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9232 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7109 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461/d983d8f1-e216-4337-915c-0ffd1d1146ca.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461/d983d8f1-e216-4337-915c-0ffd1d1146ca.json new file mode 100644 index 0000000000000000000000000000000000000000..3e2e145f4285dfe865d9cddbbaa2015e7d9278d9 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461/d983d8f1-e216-4337-915c-0ffd1d1146ca.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__2__1744311461", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7189 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7305 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9374 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7475 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780/866431a6-441f-466b-8e6f-8c5d6cf5afe8.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780/866431a6-441f-466b-8e6f-8c5d6cf5afe8.json new file mode 100644 index 0000000000000000000000000000000000000000..d2f36f692e01f6a8a639e549ed360fc9049ad169 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780/866431a6-441f-466b-8e6f-8c5d6cf5afe8.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_1_skyworkstulufull__3__1744311780", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7172 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7242 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4313 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.897 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7555 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2__1__1743896489/b8839c23-6230-4903-a2c2-f29206200c8e.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2__1__1743896489/b8839c23-6230-4903-a2c2-f29206200c8e.json new file mode 100644 index 0000000000000000000000000000000000000000..f4f9d5aa6a8ba4fa7164f645e4f36ade4311de2c --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2__1__1743896489/b8839c23-6230-4903-a2c2-f29206200c8e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_2__1__1743896489/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_2__1__1743896489", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_2__1__1743896489", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6813 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7137 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8644 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6781 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713/ae48e839-4bf6-47f5-a62c-dab942227b40.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713/ae48e839-4bf6-47f5-a62c-dab942227b40.json new file mode 100644 index 0000000000000000000000000000000000000000..7691653a93f969494eb5257aeca4645e05998681 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713/ae48e839-4bf6-47f5-a62c-dab942227b40.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713", + "id": "allenai/open_instruct_dev-rm_llama_3e-6_2_skyworkstulufull__1__1743800713", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7209 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7116 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9067 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9172 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7414 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911/262583de-511b-48b3-a240-515991d80488.json b/data/rewardbench/allenai/open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911/262583de-511b-48b3-a240-515991d80488.json new file mode 100644 index 0000000000000000000000000000000000000000..23c736763acd4e5ca485dbc9b24998884116f836 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911/262583de-511b-48b3-a240-515991d80488.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911", + "id": "allenai/open_instruct_dev-rm_llama_4e-6_1_skyworkstulufull__1__1743893911", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7266 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7347 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4313 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6339 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.897 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7697 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412/bcdd6295-d3f9-4151-afd6-29265826155e.json b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412/bcdd6295-d3f9-4151-afd6-29265826155e.json new file mode 100644 index 0000000000000000000000000000000000000000..29987e9345413a38c348285719ed70ebe8993477 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412/bcdd6295-d3f9-4151-afd6-29265826155e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412", + "id": "allenai/open_instruct_dev-rm_llamabase_1e-6_1_skyworkstulufull__1__1745386412", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5342 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6042 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.275 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5818 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3935 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922/beac85bc-eba4-4612-a149-4a04eff59e4d.json b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922/beac85bc-eba4-4612-a149-4a04eff59e4d.json new file mode 100644 index 0000000000000000000000000000000000000000..e31158c291415b5e1d6c14f6ebd5b5607e2e0361 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922/beac85bc-eba4-4612-a149-4a04eff59e4d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922", + "id": "allenai/open_instruct_dev-rm_llamabase_1e-6_2_skyworkstulufull__1__1745441922", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6111 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6884 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3063 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8289 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7576 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4628 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495/04c08dbd-aae5-499f-bedd-4ecddae681f9.json b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495/04c08dbd-aae5-499f-bedd-4ecddae681f9.json new file mode 100644 index 0000000000000000000000000000000000000000..45601b788ddc6f324f2eb62f0d170ba167ead871 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495/04c08dbd-aae5-499f-bedd-4ecddae681f9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495", + "id": "allenai/open_instruct_dev-rm_llamabase_2e-5_1_skyworkstulufull__1__1745386495", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5825 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6379 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.325 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5355 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4691 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507/d324592d-7156-44e2-9f30-f8e600c77d98.json b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507/d324592d-7156-44e2-9f30-f8e600c77d98.json new file mode 100644 index 0000000000000000000000000000000000000000..9111e2e9585f5c8bf050852854bdab90ad897d74 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507/d324592d-7156-44e2-9f30-f8e600c77d98.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507", + "id": "allenai/open_instruct_dev-rm_llamabase_2e-5_2_skyworkstulufull__1__1745386507", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5598 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5495 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5902 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.76 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7273 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3754 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507/6e314803-4a75-44bb-9771-0305b1d21add.json b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507/6e314803-4a75-44bb-9771-0305b1d21add.json new file mode 100644 index 0000000000000000000000000000000000000000..106c3a1019cfc6fe3e38a960b5e5d7bd4ac2d19b --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507/6e314803-4a75-44bb-9771-0305b1d21add.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507", + "id": "allenai/open_instruct_dev-rm_llamabase_3e-6_1_skyworkstulufull__1__1745386507", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6101 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6632 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6175 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7778 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7111 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5408 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917/05961f74-df2a-47eb-b369-754aff04e299.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917/05961f74-df2a-47eb-b369-754aff04e299.json new file mode 100644 index 0000000000000000000000000000000000000000..f3faef2ebf39a91d4be492d6df4e936e1822fe50 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917/05961f74-df2a-47eb-b369-754aff04e299.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917", + "id": "allenai/open_instruct_dev-rm_qwen32b_1e-6_skyworkstulufull__8__1748235917", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7185 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7305 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8545 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.804 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961/203bfc6d-99d9-405c-ad61-3b4e53d4406b.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961/203bfc6d-99d9-405c-ad61-3b4e53d4406b.json new file mode 100644 index 0000000000000000000000000000000000000000..665e8f6071b61b43b50b63e9a41603dd1bd2f83a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961/203bfc6d-99d9-405c-ad61-3b4e53d4406b.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961", + "id": "allenai/open_instruct_dev-rm_qwen32b_3e-6_skyworkstulufull__8__1748288961", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7325 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7474 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8141 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8763 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830/4fabf7fd-37d4-4942-8a74-1bb652116879.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830/4fabf7fd-37d4-4942-8a74-1bb652116879.json new file mode 100644 index 0000000000000000000000000000000000000000..de353a74957d84c4d29efda9aec0ec32ac052d69 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830/4fabf7fd-37d4-4942-8a74-1bb652116879.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830", + "id": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__1__1744062830", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6022 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5284 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.325 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.694 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7556 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7616 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5486 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024/a621a6ae-4fba-4f92-a4fd-3b57b540b1aa.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024/a621a6ae-4fba-4f92-a4fd-3b57b540b1aa.json new file mode 100644 index 0000000000000000000000000000000000000000..2693ac8e7f220a740d3c46b55cd5dd5a007b9e59 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024/a621a6ae-4fba-4f92-a4fd-3b57b540b1aa.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024", + "id": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworks__2__1744576024", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5948 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5579 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6776 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.72 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7394 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5863 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914/03ef7fb6-2f80-4174-8b89-cac8d8e5bad8.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914/03ef7fb6-2f80-4174-8b89-cac8d8e5bad8.json new file mode 100644 index 0000000000000000000000000000000000000000..63945fe12193690d7685bcb13406629205e5df29 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914/03ef7fb6-2f80-4174-8b89-cac8d8e5bad8.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914", + "id": "allenai/open_instruct_dev-rm_qwen_1e-6_1_skyworkstulufull__1__1743712914", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6492 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.35 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6776 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.76 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.699 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091/679160a8-cbd2-426d-b266-b1a396c073bf.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091/679160a8-cbd2-426d-b266-b1a396c073bf.json new file mode 100644 index 0000000000000000000000000000000000000000..f41d8fbac78adbd69eb8baf81097133bc6ea7dda --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091/679160a8-cbd2-426d-b266-b1a396c073bf.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091", + "id": "allenai/open_instruct_dev-rm_qwen_2e-5_1_skyworkstulufull__1__1743713091", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6764 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6885 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8622 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.802 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6984 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829/04ae1006-56db-41bf-9320-ea31c6240762.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829/04ae1006-56db-41bf-9320-ea31c6240762.json new file mode 100644 index 0000000000000000000000000000000000000000..fe58dbbdfe5105150ab931a82c975b742f76cb4a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829/04ae1006-56db-41bf-9320-ea31c6240762.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829", + "id": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__1__1744062829", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6408 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6337 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3063 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6831 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8467 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5529 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050/eba51b18-24af-4896-a6e2-f7bf06be5ea0.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050/eba51b18-24af-4896-a6e2-f7bf06be5ea0.json new file mode 100644 index 0000000000000000000000000000000000000000..f8e45e5ae0688f833d5331fe0a4096eaa0c6ca71 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050/eba51b18-24af-4896-a6e2-f7bf06be5ea0.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050", + "id": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworks__2__1744576050", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6452 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6063 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3187 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7158 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8356 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8343 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5603 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916/553fbc24-3502-4302-9fe7-adae7ebcfdf5.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916/553fbc24-3502-4302-9fe7-adae7ebcfdf5.json new file mode 100644 index 0000000000000000000000000000000000000000..96ccd866acf1a8687ad6a1971285fde6a77c417a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916/553fbc24-3502-4302-9fe7-adae7ebcfdf5.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916", + "id": "allenai/open_instruct_dev-rm_qwen_3e-6_1_skyworkstulufull__1__1743712916", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7013 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7263 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6995 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8222 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8444 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7714 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_2__1__1743023576/d5246943-8fd8-4693-8b47-d0ffe78c1b5d.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_2__1__1743023576/d5246943-8fd8-4693-8b47-d0ffe78c1b5d.json new file mode 100644 index 0000000000000000000000000000000000000000..23325cee3eff7416929ca4e312a3d027bf274ff9 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_2__1__1743023576/d5246943-8fd8-4693-8b47-d0ffe78c1b5d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_3e-6_2__1__1743023576/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_3e-6_2__1__1743023576", + "id": "allenai/open_instruct_dev-rm_qwen_3e-6_2__1__1743023576", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6369 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6905 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3187 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6236 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_3__1__1743023619/8ded5f18-ec73-4f47-950b-f649ba16f170.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_3__1__1743023619/8ded5f18-ec73-4f47-950b-f649ba16f170.json new file mode 100644 index 0000000000000000000000000000000000000000..68fe51140fbfeaacaba9a1e156144868d12de560 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwen_3e-6_3__1__1743023619/8ded5f18-ec73-4f47-950b-f649ba16f170.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwen_3e-6_3__1__1743023619/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwen_3e-6_3__1__1743023619", + "id": "allenai/open_instruct_dev-rm_qwen_3e-6_3__1__1743023619", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6221 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.325 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7455 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5852 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583/1bf0615a-fe9b-4f55-bf64-88664023fc64.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583/1bf0615a-fe9b-4f55-bf64-88664023fc64.json new file mode 100644 index 0000000000000000000000000000000000000000..b76b6141501b72d6466d136b905a76dc0cf8be06 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583/1bf0615a-fe9b-4f55-bf64-88664023fc64.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583", + "id": "allenai/open_instruct_dev-rm_qwenbase_1e-6_1_skyworkstulufull__1__1745388583", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5735 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5895 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6889 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6727 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5823 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604/0a7229c5-09c0-4b63-9fe7-ceb3ec8b6603.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604/0a7229c5-09c0-4b63-9fe7-ceb3ec8b6603.json new file mode 100644 index 0000000000000000000000000000000000000000..56fc799dca0020effa1014f3345969a34fb56d99 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604/0a7229c5-09c0-4b63-9fe7-ceb3ec8b6603.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604", + "id": "allenai/open_instruct_dev-rm_qwenbase_1e-6_2_skyworkstulufull__1__1745388604", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6336 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6337 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3063 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6885 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7244 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.802 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6465 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738/a31536a3-9df4-4748-8613-e18ec2c7cb66.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738/a31536a3-9df4-4748-8613-e18ec2c7cb66.json new file mode 100644 index 0000000000000000000000000000000000000000..a39ce3b2d5e5787f0994a593cfad5ef93de3f195 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738/a31536a3-9df4-4748-8613-e18ec2c7cb66.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738", + "id": "allenai/open_instruct_dev-rm_qwenbase_2e-5_1_skyworkstulufull__1__1745388738", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6824 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6989 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6831 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8311 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8081 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7107 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191/6a02c34f-2f64-42b7-9459-a743f649148d.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191/6a02c34f-2f64-42b7-9459-a743f649148d.json new file mode 100644 index 0000000000000000000000000000000000000000..6ef995eb05a8cd971d705b931b557f1e5cf7ca0e --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191/6a02c34f-2f64-42b7-9459-a743f649148d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191", + "id": "allenai/open_instruct_dev-rm_qwenbase_2e-5_2_skyworkstulufull__1__1745388191", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6392 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6589 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6995 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7933 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7717 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5804 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737/5c2311d4-bd04-454d-be82-f8d578a529b9.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737/5c2311d4-bd04-454d-be82-f8d578a529b9.json new file mode 100644 index 0000000000000000000000000000000000000000..164e734b6ba519ca7d6fac4c138c304a63a65654 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737/5c2311d4-bd04-454d-be82-f8d578a529b9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737", + "id": "allenai/open_instruct_dev-rm_qwenbase_3e-6_1_skyworkstulufull__1__1745388737", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.664 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6821 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8133 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8061 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7066 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138/b6c669cf-0be5-487e-a3cf-99aeef68bb3f.json b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138/b6c669cf-0be5-487e-a3cf-99aeef68bb3f.json new file mode 100644 index 0000000000000000000000000000000000000000..43e5baf8f7964d342c024782c39272498dab101b --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138/b6c669cf-0be5-487e-a3cf-99aeef68bb3f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138", + "id": "allenai/open_instruct_dev-rm_qwenbase_3e-6_2_skyworkstulufull__1__1745388138", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6678 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6505 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6831 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7978 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8808 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6632 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_1__8__1742924455/ec6a4277-4134-4f94-915c-079b326b91cd.json b/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_1__8__1742924455/ec6a4277-4134-4f94-915c-079b326b91cd.json new file mode 100644 index 0000000000000000000000000000000000000000..0d8c056d0b660d17f148ea0b4be7586d2423de18 --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_1__8__1742924455/ec6a4277-4134-4f94-915c-079b326b91cd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_tulu3_70b_1__8__1742924455/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_tulu3_70b_1__8__1742924455", + "id": "allenai/open_instruct_dev-rm_tulu3_70b_1__8__1742924455", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6618 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7958 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.325 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8311 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6323 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7311 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_2__8__1742982964/5a6111db-efbd-41dd-8b4d-208579512eda.json b/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_2__8__1742982964/5a6111db-efbd-41dd-8b4d-208579512eda.json new file mode 100644 index 0000000000000000000000000000000000000000..3493c53e7b2ed6b1ecc9bd83e5d2007bf9cfa10a --- /dev/null +++ b/data/rewardbench/allenai/open_instruct_dev-rm_tulu3_70b_2__8__1742982964/5a6111db-efbd-41dd-8b4d-208579512eda.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/allenai_open_instruct_dev-rm_tulu3_70b_2__8__1742982964/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/open_instruct_dev-rm_tulu3_70b_2__8__1742982964", + "id": "allenai/open_instruct_dev-rm_tulu3_70b_2__8__1742982964", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6605 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7789 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6448 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8844 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6667 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6195 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-2-dpo-13b/8196111c-5e67-483e-b218-06ec71c84bcd.json b/data/rewardbench/allenai/tulu-2-dpo-13b/8196111c-5e67-483e-b218-06ec71c84bcd.json new file mode 100644 index 0000000000000000000000000000000000000000..13c3435ccd60b08f1e3b57bfd413cac548c4064a --- /dev/null +++ b/data/rewardbench/allenai/tulu-2-dpo-13b/8196111c-5e67-483e-b218-06ec71c84bcd.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-2-dpo-13b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-2-dpo-13b", + "id": "allenai/tulu-2-dpo-13b", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7368 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5833 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7946 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7323 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4947 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-2-dpo-70b/94170d3e-dc5f-4e9c-bc7f-331550377c98.json b/data/rewardbench/allenai/tulu-2-dpo-70b/94170d3e-dc5f-4e9c-bc7f-331550377c98.json new file mode 100644 index 0000000000000000000000000000000000000000..9181f4788205ff40dc4008f2e6ae7f667b449193 --- /dev/null +++ b/data/rewardbench/allenai/tulu-2-dpo-70b/94170d3e-dc5f-4e9c-bc7f-331550377c98.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-2-dpo-70b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-2-dpo-70b", + "id": "allenai/tulu-2-dpo-70b", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7621 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6053 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7407 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5278 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-2-dpo-7b/2df7c8e0-f914-46aa-b5ef-7b02269e076c.json b/data/rewardbench/allenai/tulu-2-dpo-7b/2df7c8e0-f914-46aa-b5ef-7b02269e076c.json new file mode 100644 index 0000000000000000000000000000000000000000..1cdff0a440127796bedeb0d2d3a18b4c14d567cf --- /dev/null +++ b/data/rewardbench/allenai/tulu-2-dpo-7b/2df7c8e0-f914-46aa-b5ef-7b02269e076c.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-2-dpo-7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-2-dpo-7b", + "id": "allenai/tulu-2-dpo-7b", + "developer": "allenai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7212 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5614 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7527 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7176 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4774 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-v2.5-13b-preference-mix-rm/80455f67-ca62-40f4-9761-1426e6818286.json b/data/rewardbench/allenai/tulu-v2.5-13b-preference-mix-rm/80455f67-ca62-40f4-9761-1426e6818286.json new file mode 100644 index 0000000000000000000000000000000000000000..a7fc04b544f5fea90ccc93fdcc1b96dd98335a05 --- /dev/null +++ b/data/rewardbench/allenai/tulu-v2.5-13b-preference-mix-rm/80455f67-ca62-40f4-9761-1426e6818286.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-v2.5-13b-preference-mix-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-v2.5-13b-preference-mix-rm", + "id": "allenai/tulu-v2.5-13b-preference-mix-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8027 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9358 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.682 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.773 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.885 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6724 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-v2.5-13b-uf-rm/1b2cf133-8bfb-4eee-a1d7-4846beeb807d.json b/data/rewardbench/allenai/tulu-v2.5-13b-uf-rm/1b2cf133-8bfb-4eee-a1d7-4846beeb807d.json new file mode 100644 index 0000000000000000000000000000000000000000..7cdea214cc4b038d382c1abc841920c369dd3f5f --- /dev/null +++ b/data/rewardbench/allenai/tulu-v2.5-13b-uf-rm/1b2cf133-8bfb-4eee-a1d7-4846beeb807d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-v2.5-13b-uf-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-v2.5-13b-uf-rm", + "id": "allenai/tulu-v2.5-13b-uf-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4806 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3939 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4232 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5554 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4737 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6326 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-v2.5-70b-preference-mix-rm/62a22799-5830-45a6-b0c6-a9ce9fed84db.json b/data/rewardbench/allenai/tulu-v2.5-70b-preference-mix-rm/62a22799-5830-45a6-b0c6-a9ce9fed84db.json new file mode 100644 index 0000000000000000000000000000000000000000..c44564a0d7c585602ba3f4b769614354a51ab625 --- /dev/null +++ b/data/rewardbench/allenai/tulu-v2.5-70b-preference-mix-rm/62a22799-5830-45a6-b0c6-a9ce9fed84db.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-v2.5-70b-preference-mix-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-v2.5-70b-preference-mix-rm", + "id": "allenai/tulu-v2.5-70b-preference-mix-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6516 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7737 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5921 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8486 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4138 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6079 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/allenai/tulu-v2.5-70b-uf-rm/e741276b-b10a-461d-b3c7-ca7dee97ca5b.json b/data/rewardbench/allenai/tulu-v2.5-70b-uf-rm/e741276b-b10a-461d-b3c7-ca7dee97ca5b.json new file mode 100644 index 0000000000000000000000000000000000000000..f7fe7231d599c98eb224ea912f9eca36fa09cec4 --- /dev/null +++ b/data/rewardbench/allenai/tulu-v2.5-70b-uf-rm/e741276b-b10a-461d-b3c7-ca7dee97ca5b.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/allenai_tulu-v2.5-70b-uf-rm/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "allenai/tulu-v2.5-70b-uf-rm", + "id": "allenai/tulu-v2.5-70b-uf-rm", + "developer": "allenai", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7398 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8659 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7171 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7014 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.757 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5757 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-3-5-sonnet-20240620/b046cb13-6d18-49ae-ad96-b34d24f98f7c.json b/data/rewardbench/anthropic/claude-3-5-sonnet-20240620/b046cb13-6d18-49ae-ad96-b34d24f98f7c.json new file mode 100644 index 0000000000000000000000000000000000000000..8d1ea81b8e43a84dc749dc12a96bfdc7a04d3644 --- /dev/null +++ b/data/rewardbench/anthropic/claude-3-5-sonnet-20240620/b046cb13-6d18-49ae-ad96-b34d24f98f7c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-3-5-sonnet-20240620/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-3-5-sonnet-20240620", + "id": "anthropic/claude-3-5-sonnet-20240620", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6466 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5284 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3875 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5683 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8519 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8697 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.674 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-3-7-sonnet-20250219/783a38ef-c369-4d56-8bc9-614316e8ebf1.json b/data/rewardbench/anthropic/claude-3-7-sonnet-20250219/783a38ef-c369-4d56-8bc9-614316e8ebf1.json new file mode 100644 index 0000000000000000000000000000000000000000..dc32989f4c0d332eff1d8bae8d3120510113b2e2 --- /dev/null +++ b/data/rewardbench/anthropic/claude-3-7-sonnet-20250219/783a38ef-c369-4d56-8bc9-614316e8ebf1.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-3-7-sonnet-20250219/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-3-7-sonnet-20250219", + "id": "anthropic/claude-3-7-sonnet-20250219", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7539 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7326 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5437 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.75 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9033 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9212 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6723 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-3-haiku-20240307/60d34dd3-40b5-4c27-81a7-bbb2331f8e48.json b/data/rewardbench/anthropic/claude-3-haiku-20240307/60d34dd3-40b5-4c27-81a7-bbb2331f8e48.json new file mode 100644 index 0000000000000000000000000000000000000000..0519bd1941bb0b7ebbf7233ade98831080f4d2c2 --- /dev/null +++ b/data/rewardbench/anthropic/claude-3-haiku-20240307/60d34dd3-40b5-4c27-81a7-bbb2331f8e48.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-3-haiku-20240307/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-3-haiku-20240307", + "id": "anthropic/claude-3-haiku-20240307", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3711 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4042 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3552 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.595 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.501 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0899 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-3-opus-20240229/78ac10cd-7cc0-4f4b-9716-30c82dd8e844.json b/data/rewardbench/anthropic/claude-3-opus-20240229/78ac10cd-7cc0-4f4b-9716-30c82dd8e844.json new file mode 100644 index 0000000000000000000000000000000000000000..3267f0e62c7c1ff69c6ecc9faf390d9d62efb6b4 --- /dev/null +++ b/data/rewardbench/anthropic/claude-3-opus-20240229/78ac10cd-7cc0-4f4b-9716-30c82dd8e844.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-3-opus-20240229/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-3-opus-20240229", + "id": "anthropic/claude-3-opus-20240229", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5744 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5389 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5137 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8378 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6646 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5601 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-opus-4-20250514/e1c143d0-fa21-4c1d-8677-511f708b1770.json b/data/rewardbench/anthropic/claude-opus-4-20250514/e1c143d0-fa21-4c1d-8677-511f708b1770.json new file mode 100644 index 0000000000000000000000000000000000000000..e308713683fa451e2e331f69f3f39067d26a36e9 --- /dev/null +++ b/data/rewardbench/anthropic/claude-opus-4-20250514/e1c143d0-fa21-4c1d-8677-511f708b1770.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-opus-4-20250514/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-opus-4-20250514", + "id": "anthropic/claude-opus-4-20250514", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7648 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8267 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7491 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8954 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8616 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8375 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/anthropic/claude-sonnet-4-20250514/bc8e10a2-c464-499f-b7d4-43af3707db6a.json b/data/rewardbench/anthropic/claude-sonnet-4-20250514/bc8e10a2-c464-499f-b7d4-43af3707db6a.json new file mode 100644 index 0000000000000000000000000000000000000000..a520e891cee62738304faf2bfff8fe62acad7017 --- /dev/null +++ b/data/rewardbench/anthropic/claude-sonnet-4-20250514/bc8e10a2-c464-499f-b7d4-43af3707db6a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/anthropic_claude-sonnet-4-20250514/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "anthropic/claude-sonnet-4-20250514", + "id": "anthropic/claude-sonnet-4-20250514", + "developer": "anthropic", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7117 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7612 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3594 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7049 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8909 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7939 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/berkeley-nest/Starling-RM-7B-alpha/8ad5c679-7bca-4f4a-928f-fa3e0315f814.json b/data/rewardbench/berkeley-nest/Starling-RM-7B-alpha/8ad5c679-7bca-4f4a-928f-fa3e0315f814.json new file mode 100644 index 0000000000000000000000000000000000000000..ea480ed9e2dbf5af0256c452c89217c829f8666d --- /dev/null +++ b/data/rewardbench/berkeley-nest/Starling-RM-7B-alpha/8ad5c679-7bca-4f4a-928f-fa3e0315f814.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/berkeley-nest_Starling-RM-7B-alpha/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "berkeley-nest/Starling-RM-7B-alpha", + "id": "berkeley-nest/Starling-RM-7B-alpha", + "developer": "berkeley-nest", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7113 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9804 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4561 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.58 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6794 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/cohere/Cohere March 2024/486e23bb-a086-47ec-bf5c-0d224b80442e.json b/data/rewardbench/cohere/Cohere March 2024/486e23bb-a086-47ec-bf5c-0d224b80442e.json new file mode 100644 index 0000000000000000000000000000000000000000..63da81b687e427030bc1868d94d43b177be4c374 --- /dev/null +++ b/data/rewardbench/cohere/Cohere March 2024/486e23bb-a086-47ec-bf5c-0d224b80442e.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/cohere_Cohere March 2024/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Cohere March 2024", + "id": "cohere/Cohere March 2024", + "developer": "cohere", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8511 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9469 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6513 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.877 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9817 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7458 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/cohere/Cohere May 2024/5889db5c-6146-4237-9edd-454e3ae6a387.json b/data/rewardbench/cohere/Cohere May 2024/5889db5c-6146-4237-9edd-454e3ae6a387.json new file mode 100644 index 0000000000000000000000000000000000000000..1b96669afad94b637e5d44b477a0a616f0fc6da0 --- /dev/null +++ b/data/rewardbench/cohere/Cohere May 2024/5889db5c-6146-4237-9edd-454e3ae6a387.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/cohere_Cohere May 2024/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "Cohere May 2024", + "id": "cohere/Cohere May 2024", + "developer": "cohere", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8816 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7127 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.923 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9768 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.782 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/facebook/Self-taught-Llama-3-70B/89c9dc02-5265-4f61-9881-b66653565830.json b/data/rewardbench/facebook/Self-taught-Llama-3-70B/89c9dc02-5265-4f61-9881-b66653565830.json new file mode 100644 index 0000000000000000000000000000000000000000..d60e4dac4db4fa9895c6c11ea6663f88626bcc95 --- /dev/null +++ b/data/rewardbench/facebook/Self-taught-Llama-3-70B/89c9dc02-5265-4f61-9881-b66653565830.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/facebook_Self-taught-Llama-3-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "facebook/Self-taught-Llama-3-70B", + "id": "facebook/Self-taught-Llama-3-70B", + "developer": "facebook", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8863 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8399 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9108 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8251 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/facebook/Self-taught-evaluator-llama3.1-70B/70baebf8-cb59-4a57-a5e4-0c514e6e1513.json b/data/rewardbench/facebook/Self-taught-evaluator-llama3.1-70B/70baebf8-cb59-4a57-a5e4-0c514e6e1513.json new file mode 100644 index 0000000000000000000000000000000000000000..bb99b0c03fe6eceb2692ca71fab16715aa865f2b --- /dev/null +++ b/data/rewardbench/facebook/Self-taught-evaluator-llama3.1-70B/70baebf8-cb59-4a57-a5e4-0c514e6e1513.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/facebook_Self-taught-evaluator-llama3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "facebook/Self-taught-evaluator-llama3.1-70B", + "id": "facebook/Self-taught-evaluator-llama3.1-70B", + "developer": "facebook", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9001 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8509 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8959 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8844 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/general-preference/GPM-Gemma-2B/cac0200c-27da-4141-8481-c8e6fc3c8c90.json b/data/rewardbench/general-preference/GPM-Gemma-2B/cac0200c-27da-4141-8481-c8e6fc3c8c90.json new file mode 100644 index 0000000000000000000000000000000000000000..b816b2d06bc832f1e566df518c128c11d7b68263 --- /dev/null +++ b/data/rewardbench/general-preference/GPM-Gemma-2B/cac0200c-27da-4141-8481-c8e6fc3c8c90.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/general-preference_GPM-Gemma-2B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "general-preference/GPM-Gemma-2B", + "id": "general-preference/GPM-Gemma-2B", + "developer": "general-preference", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7449 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7151 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6974 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8122 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.755 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/general-preference/GPM-Llama-3.1-8B/e4e30e1b-60d9-4673-83b9-a9b8caa0c57d.json b/data/rewardbench/general-preference/GPM-Llama-3.1-8B/e4e30e1b-60d9-4673-83b9-a9b8caa0c57d.json new file mode 100644 index 0000000000000000000000000000000000000000..21492e51bff203c9183afc2879fb56302749c79e --- /dev/null +++ b/data/rewardbench/general-preference/GPM-Llama-3.1-8B/e4e30e1b-60d9-4673-83b9-a9b8caa0c57d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/general-preference_GPM-Llama-3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "general-preference/GPM-Llama-3.1-8B", + "id": "general-preference/GPM-Llama-3.1-8B", + "developer": "general-preference", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9224 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.933 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.886 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9108 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9597 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/flame-1.0-24B-july-2024/50db9a67-77a0-433f-b8e8-462cdb3925d3.json b/data/rewardbench/google/flame-1.0-24B-july-2024/50db9a67-77a0-433f-b8e8-462cdb3925d3.json new file mode 100644 index 0000000000000000000000000000000000000000..24a29648d34896ae073a2d1a19e3261352e63695 --- /dev/null +++ b/data/rewardbench/google/flame-1.0-24B-july-2024/50db9a67-77a0-433f-b8e8-462cdb3925d3.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_flame-1.0-24B-july-2024/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/flame-1.0-24B-july-2024", + "id": "google/flame-1.0-24B-july-2024", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8781 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7566 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8959 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.938 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-1.5-flash-001/604f5b90-294c-4080-b4a0-8d554166d8cd.json b/data/rewardbench/google/gemini-1.5-flash-001/604f5b90-294c-4080-b4a0-8d554166d8cd.json new file mode 100644 index 0000000000000000000000000000000000000000..3b8ab94bc40ae5929fceafe3463e14fb6de4d895 --- /dev/null +++ b/data/rewardbench/google/gemini-1.5-flash-001/604f5b90-294c-4080-b4a0-8d554166d8cd.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_gemini-1.5-flash-001/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-1.5-flash-001", + "id": "google/gemini-1.5-flash-001", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8054 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6349 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8696 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8512 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6937 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-1.5-flash-8b/7ba58d8a-c106-439e-9730-57b7b09419e4.json b/data/rewardbench/google/gemini-1.5-flash-8b/7ba58d8a-c106-439e-9730-57b7b09419e4.json new file mode 100644 index 0000000000000000000000000000000000000000..586da67de4c11957d48dbea52ccec7eaf849cb2e --- /dev/null +++ b/data/rewardbench/google/gemini-1.5-flash-8b/7ba58d8a-c106-439e-9730-57b7b09419e4.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/google_gemini-1.5-flash-8b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-1.5-flash-8b", + "id": "google/gemini-1.5-flash-8b", + "developer": "google", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4851 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4611 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5082 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6622 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6747 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2421 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-1.5-flash-8b/b6faa66f-6990-466f-86e8-0bccb9495233.json b/data/rewardbench/google/gemini-1.5-flash-8b/b6faa66f-6990-466f-86e8-0bccb9495233.json new file mode 100644 index 0000000000000000000000000000000000000000..6c7fee497c39d39a5397824c9a63194b889557dd --- /dev/null +++ b/data/rewardbench/google/gemini-1.5-flash-8b/b6faa66f-6990-466f-86e8-0bccb9495233.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_gemini-1.5-flash-8b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "gemini-1.5-flash-8b", + "id": "google/gemini-1.5-flash-8b", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7601 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5987 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7399 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7575 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-1.5-pro-0514/f7adee07-81d2-463e-9b32-f84d5eee77a8.json b/data/rewardbench/google/gemini-1.5-pro-0514/f7adee07-81d2-463e-9b32-f84d5eee77a8.json new file mode 100644 index 0000000000000000000000000000000000000000..175ef548f38f90dbc2c05190fe3fd93dcfb647a7 --- /dev/null +++ b/data/rewardbench/google/gemini-1.5-pro-0514/f7adee07-81d2-463e-9b32-f84d5eee77a8.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_gemini-1.5-pro-0514/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-1.5-pro-0514", + "id": "google/gemini-1.5-pro-0514", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.882 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9232 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8059 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8791 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9199 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-1.5-pro-0924/2fe52b14-8dc1-46b0-b2e3-5363a3de3588.json b/data/rewardbench/google/gemini-1.5-pro-0924/2fe52b14-8dc1-46b0-b2e3-5363a3de3588.json new file mode 100644 index 0000000000000000000000000000000000000000..1b983cfa37704c832dd82b232fe73faeac6e90aa --- /dev/null +++ b/data/rewardbench/google/gemini-1.5-pro-0924/2fe52b14-8dc1-46b0-b2e3-5363a3de3588.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_gemini-1.5-pro-0924/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-1.5-pro-0924", + "id": "google/gemini-1.5-pro-0924", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8678 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9413 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7697 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8581 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9022 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-2.5-flash-preview-04-17/c3e737c3-98e0-4b31-9ba0-e881b34d45a9.json b/data/rewardbench/google/gemini-2.5-flash-preview-04-17/c3e737c3-98e0-4b31-9ba0-e881b34d45a9.json new file mode 100644 index 0000000000000000000000000000000000000000..0dad38ce30d89de6ca173c5b0be7e887d972aed1 --- /dev/null +++ b/data/rewardbench/google/gemini-2.5-flash-preview-04-17/c3e737c3-98e0-4b31-9ba0-e881b34d45a9.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/google_gemini-2.5-flash-preview-04-17/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-2.5-flash-preview-04-17", + "id": "google/gemini-2.5-flash-preview-04-17", + "developer": "google", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7721 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6574 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5531 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8115 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9094 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8672 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8341 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-2.5-flash/269caf81-718b-4793-86e5-d421c689337c.json b/data/rewardbench/google/gemini-2.5-flash/269caf81-718b-4793-86e5-d421c689337c.json new file mode 100644 index 0000000000000000000000000000000000000000..95f1cfa95d16856ed10f5c656b0dd7110a8eca49 --- /dev/null +++ b/data/rewardbench/google/gemini-2.5-flash/269caf81-718b-4793-86e5-d421c689337c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/google_gemini-2.5-flash/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-2.5-flash", + "id": "google/gemini-2.5-flash", + "developer": "google", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7767 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.674 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.575 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.852 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.909 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.841 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.809 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-2.5-pro-preview-05-06/53934cd8-f355-4449-a198-8a95c1d8bb21.json b/data/rewardbench/google/gemini-2.5-pro-preview-05-06/53934cd8-f355-4449-a198-8a95c1d8bb21.json new file mode 100644 index 0000000000000000000000000000000000000000..f438621aac9731d03810dbd28c69df25e6fb0183 --- /dev/null +++ b/data/rewardbench/google/gemini-2.5-pro-preview-05-06/53934cd8-f355-4449-a198-8a95c1d8bb21.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/google_gemini-2.5-pro-preview-05-06/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-2.5-pro-preview-05-06", + "id": "google/gemini-2.5-pro-preview-05-06", + "developer": "google", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6775 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6532 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4688 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5342 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8806 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8308 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6973 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemini-2.5-pro/b6e9a1c8-bad6-41bc-9a3e-8f6668b4e03e.json b/data/rewardbench/google/gemini-2.5-pro/b6e9a1c8-bad6-41bc-9a3e-8f6668b4e03e.json new file mode 100644 index 0000000000000000000000000000000000000000..d6fb9023b4e3f6596ce8edb9a645f4368fafef02 --- /dev/null +++ b/data/rewardbench/google/gemini-2.5-pro/b6e9a1c8-bad6-41bc-9a3e-8f6668b4e03e.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/google_gemini-2.5-pro/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemini-2.5-pro", + "id": "google/gemini-2.5-pro", + "developer": "google", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7948 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.755 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.619 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.898 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.881 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.805 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.811 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/google/gemma-2-27b-it/f4a5ab9d-4dfc-4b19-8796-339cf1dcabee.json b/data/rewardbench/google/gemma-2-27b-it/f4a5ab9d-4dfc-4b19-8796-339cf1dcabee.json new file mode 100644 index 0000000000000000000000000000000000000000..9de746a6e973842afff59ef3fec527b8f7ae8211 --- /dev/null +++ b/data/rewardbench/google/gemma-2-27b-it/f4a5ab9d-4dfc-4b19-8796-339cf1dcabee.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/google_gemma-2-27b-it/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "google/gemma-2-27b-it", + "id": "google/gemma-2-27b-it", + "developer": "google", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.809 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9483 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.591 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8635 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.833 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/2b8563e7-9d58-428b-a16f-6394fc84bf3c.json b/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/2b8563e7-9d58-428b-a16f-6394fc84bf3c.json new file mode 100644 index 0000000000000000000000000000000000000000..355f2016f16883cbfba87b784ffa5fa096957614 --- /dev/null +++ b/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/2b8563e7-9d58-428b-a16f-6394fc84bf3c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/hendrydong_Mistral-RM-for-RAFT-GSHF-v0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "hendrydong/Mistral-RM-for-RAFT-GSHF-v0", + "id": "hendrydong/Mistral-RM-for-RAFT-GSHF-v0", + "developer": "hendrydong", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5851 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5779 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6011 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6956 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6747 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5988 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/deccf1b3-3e41-4945-b063-e5d315649446.json b/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/deccf1b3-3e41-4945-b063-e5d315649446.json new file mode 100644 index 0000000000000000000000000000000000000000..b456cf0d44a7ae45db11fad31da2d60383953a07 --- /dev/null +++ b/data/rewardbench/hendrydong/Mistral-RM-for-RAFT-GSHF-v0/deccf1b3-3e41-4945-b063-e5d315649446.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/hendrydong_Mistral-RM-for-RAFT-GSHF-v0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "hendrydong/Mistral-RM-for-RAFT-GSHF-v0", + "id": "hendrydong/Mistral-RM-for-RAFT-GSHF-v0", + "developer": "hendrydong", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7847 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9832 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5789 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.85 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7434 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7508 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/infly/INF-ORM-Llama3.1-70B/24003f1d-31f0-4735-b762-144f71025957.json b/data/rewardbench/infly/INF-ORM-Llama3.1-70B/24003f1d-31f0-4735-b762-144f71025957.json new file mode 100644 index 0000000000000000000000000000000000000000..4d39cffbf60b053d609bb7d01805dc4e55fc182e --- /dev/null +++ b/data/rewardbench/infly/INF-ORM-Llama3.1-70B/24003f1d-31f0-4735-b762-144f71025957.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/infly_INF-ORM-Llama3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "infly/INF-ORM-Llama3.1-70B", + "id": "infly/INF-ORM-Llama3.1-70B", + "developer": "infly", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9511 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9101 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9365 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9912 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/infly/INF-ORM-Llama3.1-70B/c0c64cfb-7c3d-49c5-ac49-bb426f004f48.json b/data/rewardbench/infly/INF-ORM-Llama3.1-70B/c0c64cfb-7c3d-49c5-ac49-bb426f004f48.json new file mode 100644 index 0000000000000000000000000000000000000000..f4dc92c3ad4bf06b7c5bf2225111ae120b684b7c --- /dev/null +++ b/data/rewardbench/infly/INF-ORM-Llama3.1-70B/c0c64cfb-7c3d-49c5-ac49-bb426f004f48.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/infly_INF-ORM-Llama3.1-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "infly/INF-ORM-Llama3.1-70B", + "id": "infly/INF-ORM-Llama3.1-70B", + "developer": "infly", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7648 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7411 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6995 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9644 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.903 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8622 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-1_8b-reward/560b7027-41fc-43af-8094-0e69f42a152c.json b/data/rewardbench/internlm/internlm2-1_8b-reward/560b7027-41fc-43af-8094-0e69f42a152c.json new file mode 100644 index 0000000000000000000000000000000000000000..1d60d1de49871fff09148c489759cd06c3070467 --- /dev/null +++ b/data/rewardbench/internlm/internlm2-1_8b-reward/560b7027-41fc-43af-8094-0e69f42a152c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/internlm_internlm2-1_8b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-1_8b-reward", + "id": "internlm/internlm2-1_8b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3902 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2758 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4426 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4711 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.596 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1934 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-1_8b-reward/e829c7ed-baa8-4c77-a283-623e6e0bad1d.json b/data/rewardbench/internlm/internlm2-1_8b-reward/e829c7ed-baa8-4c77-a283-623e6e0bad1d.json new file mode 100644 index 0000000000000000000000000000000000000000..0bc2859e733a49fe98a975ee969111e4ffe99d89 --- /dev/null +++ b/data/rewardbench/internlm/internlm2-1_8b-reward/e829c7ed-baa8-4c77-a283-623e6e0bad1d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/internlm_internlm2-1_8b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-1_8b-reward", + "id": "internlm/internlm2-1_8b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8217 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9358 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8162 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8724 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-20b-reward/09e959b6-adff-4de1-be3f-85719d09f34b.json b/data/rewardbench/internlm/internlm2-20b-reward/09e959b6-adff-4de1-be3f-85719d09f34b.json new file mode 100644 index 0000000000000000000000000000000000000000..e364b003ee67b1f9cd15dd45be22148569864f9c --- /dev/null +++ b/data/rewardbench/internlm/internlm2-20b-reward/09e959b6-adff-4de1-be3f-85719d09f34b.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/internlm_internlm2-20b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-20b-reward", + "id": "internlm/internlm2-20b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9016 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9888 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7654 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8946 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9576 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-20b-reward/1fcf3142-b2ab-4f78-8696-2328bcf5f598.json b/data/rewardbench/internlm/internlm2-20b-reward/1fcf3142-b2ab-4f78-8696-2328bcf5f598.json new file mode 100644 index 0000000000000000000000000000000000000000..06a1dffc5685e0409b67f5105fcc90f1a4c04908 --- /dev/null +++ b/data/rewardbench/internlm/internlm2-20b-reward/1fcf3142-b2ab-4f78-8696-2328bcf5f598.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/internlm_internlm2-20b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-20b-reward", + "id": "internlm/internlm2-20b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5628 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5558 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3625 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5738 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6111 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7253 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5483 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-7b-reward/2e703293-54aa-4cce-81ab-6fb151001abd.json b/data/rewardbench/internlm/internlm2-7b-reward/2e703293-54aa-4cce-81ab-6fb151001abd.json new file mode 100644 index 0000000000000000000000000000000000000000..a50189caa3da8262cc39056915e4f5c144122321 --- /dev/null +++ b/data/rewardbench/internlm/internlm2-7b-reward/2e703293-54aa-4cce-81ab-6fb151001abd.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/internlm_internlm2-7b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-7b-reward", + "id": "internlm/internlm2-7b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5335 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4211 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5628 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5956 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5164 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/internlm/internlm2-7b-reward/4bf01d2d-c813-4200-b61f-1d2383871db0.json b/data/rewardbench/internlm/internlm2-7b-reward/4bf01d2d-c813-4200-b61f-1d2383871db0.json new file mode 100644 index 0000000000000000000000000000000000000000..4a8e163e543e8f8773d3679e97995d5dd7043e3b --- /dev/null +++ b/data/rewardbench/internlm/internlm2-7b-reward/4bf01d2d-c813-4200-b61f-1d2383871db0.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/internlm_internlm2-7b-reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "internlm/internlm2-7b-reward", + "id": "internlm/internlm2-7b-reward", + "developer": "internlm", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8759 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9916 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6952 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8716 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9453 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/jondurbin/bagel-dpo-34b-v0.5/e8f881c8-4829-46b6-adf8-d67fc705fd4c.json b/data/rewardbench/jondurbin/bagel-dpo-34b-v0.5/e8f881c8-4829-46b6-adf8-d67fc705fd4c.json new file mode 100644 index 0000000000000000000000000000000000000000..363ff957bee112eee2ba25035765c0393176f953 --- /dev/null +++ b/data/rewardbench/jondurbin/bagel-dpo-34b-v0.5/e8f881c8-4829-46b6-adf8-d67fc705fd4c.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/jondurbin_bagel-dpo-34b-v0.5/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "jondurbin/bagel-dpo-34b-v0.5", + "id": "jondurbin/bagel-dpo-34b-v0.5", + "developer": "jondurbin", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7215 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9385 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5504 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8889 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4487 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/llm-blender/PairRM-hf/90e40317-51de-4b76-913b-cd75cd839237.json b/data/rewardbench/llm-blender/PairRM-hf/90e40317-51de-4b76-913b-cd75cd839237.json new file mode 100644 index 0000000000000000000000000000000000000000..34d447ec178e4d856f8f10c55104aa432bf67fae --- /dev/null +++ b/data/rewardbench/llm-blender/PairRM-hf/90e40317-51de-4b76-913b-cd75cd839237.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/llm-blender_PairRM-hf/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "llm-blender/PairRM-hf", + "id": "llm-blender/PairRM-hf", + "developer": "llm-blender", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6087 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9022 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5219 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.477 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4898 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6961 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/mattshumer/Reflection-70B/d987d22b-d4e5-48c4-a026-90bbd9f0db04.json b/data/rewardbench/mattshumer/Reflection-70B/d987d22b-d4e5-48c4-a026-90bbd9f0db04.json new file mode 100644 index 0000000000000000000000000000000000000000..c9495543383bc2624930b85fc32eff8a07e41719 --- /dev/null +++ b/data/rewardbench/mattshumer/Reflection-70B/d987d22b-d4e5-48c4-a026-90bbd9f0db04.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/mattshumer_Reflection-70B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "mattshumer/Reflection-70B", + "id": "mattshumer/Reflection-70B", + "developer": "mattshumer", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8422 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7061 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8318 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8562 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3-70B-Instruct/64d658ee-5e9d-426e-9868-12b94af12511.json b/data/rewardbench/meta-llama/Meta-Llama-3-70B-Instruct/64d658ee-5e9d-426e-9868-12b94af12511.json new file mode 100644 index 0000000000000000000000000000000000000000..e71e3897d55031b94e4d5c7e635b5e476e3ca920 --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3-70B-Instruct/64d658ee-5e9d-426e-9868-12b94af12511.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3-70B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3-70B-Instruct", + "id": "meta-llama/Meta-Llama-3-70B-Instruct", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7627 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9763 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5888 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7297 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7854 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7035 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3-8B-Instruct/7393c63b-d32a-4744-9340-4b9898882130.json b/data/rewardbench/meta-llama/Meta-Llama-3-8B-Instruct/7393c63b-d32a-4744-9340-4b9898882130.json new file mode 100644 index 0000000000000000000000000000000000000000..cc46bc2ebc685a3a7be60a4a55b7c76c3475944b --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3-8B-Instruct/7393c63b-d32a-4744-9340-4b9898882130.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3-8B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3-8B-Instruct", + "id": "meta-llama/Meta-Llama-3-8B-Instruct", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.645 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8547 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4156 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6797 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6482 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6082 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo/a4d04c01-c469-4768-b4cd-25e803f12f35.json b/data/rewardbench/meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo/a4d04c01-c469-4768-b4cd-25e803f12f35.json new file mode 100644 index 0000000000000000000000000000000000000000..d80adc5576edae71782e8e8a31ed255f9e71c57b --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo/a4d04c01-c469-4768-b4cd-25e803f12f35.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3.1-405B-Instruct-Turbo/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + "id": "meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8412 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9721 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7456 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8715 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo/e279aa69-5141-4287-b4d3-e19fed31dd2d.json b/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo/e279aa69-5141-4287-b4d3-e19fed31dd2d.json new file mode 100644 index 0000000000000000000000000000000000000000..d364fc4474f2db77d6ffd4412a84e239aa9ff14a --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo/e279aa69-5141-4287-b4d3-e19fed31dd2d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3.1-70B-Instruct-Turbo/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + "id": "meta-llama/Meta-Llama-3.1-70B-Instruct-Turbo", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7808 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8757 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6689 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7507 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.828 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct/3d2dcaec-a1ae-4d0d-8629-a94ba351d3a8.json b/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct/3d2dcaec-a1ae-4d0d-8629-a94ba351d3a8.json new file mode 100644 index 0000000000000000000000000000000000000000..d802391269bc750400cfb5e7e73aafdbe1dec87c --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3.1-70B-Instruct/3d2dcaec-a1ae-4d0d-8629-a94ba351d3a8.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3.1-70B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3.1-70B-Instruct", + "id": "meta-llama/Meta-Llama-3.1-70B-Instruct", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8405 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9721 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7018 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8284 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8599 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo/5979a8de-fb22-4158-b079-7a8bdbc5971b.json b/data/rewardbench/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo/5979a8de-fb22-4158-b079-7a8bdbc5971b.json new file mode 100644 index 0000000000000000000000000000000000000000..4a6d43713e753caa34b978d8de9477fda86f098c --- /dev/null +++ b/data/rewardbench/meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo/5979a8de-fb22-4158-b079-7a8bdbc5971b.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-llama_Meta-Llama-3.1-8B-Instruct-Turbo/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + "id": "meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo", + "developer": "meta-llama", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6565 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8073 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4978 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6399 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6811 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/meta-metrics/MetaMetrics-RM-v1.0/58a28566-bad7-473f-9da1-bc356d48c774.json b/data/rewardbench/meta-metrics/MetaMetrics-RM-v1.0/58a28566-bad7-473f-9da1-bc356d48c774.json new file mode 100644 index 0000000000000000000000000000000000000000..83fd92aa9759b8d9b412ce76d9d53304b6feca43 --- /dev/null +++ b/data/rewardbench/meta-metrics/MetaMetrics-RM-v1.0/58a28566-bad7-473f-9da1-bc356d48c774.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/meta-metrics_MetaMetrics-RM-v1.0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "meta-metrics/MetaMetrics-RM-v1.0", + "id": "meta-metrics/MetaMetrics-RM-v1.0", + "developer": "meta-metrics", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9342 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9832 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.864 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9081 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9816 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/mightbe/Better-PairRM/bda8e2f6-1026-4f9e-85f9-b62bafec3657.json b/data/rewardbench/mightbe/Better-PairRM/bda8e2f6-1026-4f9e-85f9-b62bafec3657.json new file mode 100644 index 0000000000000000000000000000000000000000..1afd7414c75b86ace4fae9c69eb189a9d5ab0056 --- /dev/null +++ b/data/rewardbench/mightbe/Better-PairRM/bda8e2f6-1026-4f9e-85f9-b62bafec3657.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/mightbe_Better-PairRM/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "mightbe/Better-PairRM", + "id": "mightbe/Better-PairRM", + "developer": "mightbe", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.673 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3925 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8203 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4983 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.724 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/mistralai/Mixtral-8x7B-Instruct-v0.1/36ca19ff-8e2e-4f3a-91cd-5d531fec80fa.json b/data/rewardbench/mistralai/Mixtral-8x7B-Instruct-v0.1/36ca19ff-8e2e-4f3a-91cd-5d531fec80fa.json new file mode 100644 index 0000000000000000000000000000000000000000..1742be1083591a4a89a3647892cffd3ec0a0d1b8 --- /dev/null +++ b/data/rewardbench/mistralai/Mixtral-8x7B-Instruct-v0.1/36ca19ff-8e2e-4f3a-91cd-5d531fec80fa.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/mistralai_Mixtral-8x7B-Instruct-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "mistralai/Mixtral-8x7B-Instruct-v0.1", + "id": "mistralai/Mixtral-8x7B-Instruct-v0.1", + "developer": "mistralai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7455 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9497 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6404 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7257 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7872 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5033 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/my_model/f4bea180-9796-48e3-bd74-4f2f16897b14.json b/data/rewardbench/my_model/f4bea180-9796-48e3-bd74-4f2f16897b14.json new file mode 100644 index 0000000000000000000000000000000000000000..012a1c56064dc3bc2df442cfa0015a42c813a741 --- /dev/null +++ b/data/rewardbench/my_model/f4bea180-9796-48e3-bd74-4f2f16897b14.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/my_model_/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "my_model/", + "id": "my_model/", + "developer": "my_model", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5267 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4553 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5592 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4392 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6532 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Gemma-2-27B/2684d984-b0b9-4aa8-a7df-4c0daf392083.json b/data/rewardbench/nicolinho/QRM-Gemma-2-27B/2684d984-b0b9-4aa8-a7df-4c0daf392083.json new file mode 100644 index 0000000000000000000000000000000000000000..749136f3e83dc20985515fc25cf2de9c98bb6ada --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Gemma-2-27B/2684d984-b0b9-4aa8-a7df-4c0daf392083.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nicolinho_QRM-Gemma-2-27B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Gemma-2-27B", + "id": "nicolinho/QRM-Gemma-2-27B", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9444 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9013 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.927 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9826 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Gemma-2-27B/52cd7dee-06c9-4a05-a5b7-54dfab2a8b5f.json b/data/rewardbench/nicolinho/QRM-Gemma-2-27B/52cd7dee-06c9-4a05-a5b7-54dfab2a8b5f.json new file mode 100644 index 0000000000000000000000000000000000000000..eb08d98b834e644079a4f8d7f0c36e83e3dfe8b6 --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Gemma-2-27B/52cd7dee-06c9-4a05-a5b7-54dfab2a8b5f.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/nicolinho_QRM-Gemma-2-27B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Gemma-2-27B", + "id": "nicolinho/QRM-Gemma-2-27B", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7667 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7853 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3719 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6995 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9578 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9535 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8321 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Llama3-8B/967ad03e-7026-41bd-8a9c-0bb2d608bf2b.json b/data/rewardbench/nicolinho/QRM-Llama3-8B/967ad03e-7026-41bd-8a9c-0bb2d608bf2b.json new file mode 100644 index 0000000000000000000000000000000000000000..a9d69f6d71fe352b29f75fbb5cdf76e746ac8b5c --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Llama3-8B/967ad03e-7026-41bd-8a9c-0bb2d608bf2b.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nicolinho_QRM-Llama3-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Llama3-8B", + "id": "nicolinho/QRM-Llama3-8B", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.911 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8114 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8986 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9758 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/369a086b-25be-4550-857a-eddf2a7b7b67.json b/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/369a086b-25be-4550-857a-eddf2a7b7b67.json new file mode 100644 index 0000000000000000000000000000000000000000..c13172cf3b6987148d639a92c39b1f5264288931 --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/369a086b-25be-4550-857a-eddf2a7b7b67.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nicolinho_QRM-Llama3.1-8B-v2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Llama3.1-8B-v2", + "id": "nicolinho/QRM-Llama3.1-8B-v2", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9314 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8684 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9257 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9677 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/61408768-3cd3-48f8-937d-5697ac7e4753.json b/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/61408768-3cd3-48f8-937d-5697ac7e4753.json new file mode 100644 index 0000000000000000000000000000000000000000..f38bcc4639bb9d93460fbd6b13417d04a7037a8c --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Llama3.1-8B-v2/61408768-3cd3-48f8-937d-5697ac7e4753.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/nicolinho_QRM-Llama3.1-8B-v2/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Llama3.1-8B-v2", + "id": "nicolinho/QRM-Llama3.1-8B-v2", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7074 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6653 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4062 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.612 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9467 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8909 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7234 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nicolinho/QRM-Llama3.1-8B/8c31f2db-f1cc-4bfa-9572-83dc88f5d17a.json b/data/rewardbench/nicolinho/QRM-Llama3.1-8B/8c31f2db-f1cc-4bfa-9572-83dc88f5d17a.json new file mode 100644 index 0000000000000000000000000000000000000000..f3d1caf980b8c59d8b68c2c43bcec3957693f75c --- /dev/null +++ b/data/rewardbench/nicolinho/QRM-Llama3.1-8B/8c31f2db-f1cc-4bfa-9572-83dc88f5d17a.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nicolinho_QRM-Llama3.1-8B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nicolinho/QRM-Llama3.1-8B", + "id": "nicolinho/QRM-Llama3.1-8B", + "developer": "nicolinho", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9306 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8969 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.923 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9583 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nvidia/Llama-3.1-Nemotron-70B-Reward/73f5beca-afd3-447a-8534-3a11ba6289be.json b/data/rewardbench/nvidia/Llama-3.1-Nemotron-70B-Reward/73f5beca-afd3-447a-8534-3a11ba6289be.json new file mode 100644 index 0000000000000000000000000000000000000000..04ce2d9698b3109e1a26cbb16c43f46bd86e5ed8 --- /dev/null +++ b/data/rewardbench/nvidia/Llama-3.1-Nemotron-70B-Reward/73f5beca-afd3-447a-8534-3a11ba6289be.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nvidia_Llama-3.1-Nemotron-70B-Reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nvidia/Llama-3.1-Nemotron-70B-Reward", + "id": "nvidia/Llama-3.1-Nemotron-70B-Reward", + "developer": "nvidia", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9411 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8575 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9514 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9807 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nvidia/Llama3-70B-SteerLM-RM/1c6b072b-a546-402d-b9d4-d9f2255733f0.json b/data/rewardbench/nvidia/Llama3-70B-SteerLM-RM/1c6b072b-a546-402d-b9d4-d9f2255733f0.json new file mode 100644 index 0000000000000000000000000000000000000000..44b1b9b81a2e7796778fec675c2b420d0845e05b --- /dev/null +++ b/data/rewardbench/nvidia/Llama3-70B-SteerLM-RM/1c6b072b-a546-402d-b9d4-d9f2255733f0.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nvidia_Llama3-70B-SteerLM-RM/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nvidia/Llama3-70B-SteerLM-RM", + "id": "nvidia/Llama3-70B-SteerLM-RM", + "developer": "nvidia", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8877 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9134 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8026 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9284 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9064 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/nvidia/Nemotron-4-340B-Reward/fd38add4-a301-4bae-a0f5-63a59e979d7a.json b/data/rewardbench/nvidia/Nemotron-4-340B-Reward/fd38add4-a301-4bae-a0f5-63a59e979d7a.json new file mode 100644 index 0000000000000000000000000000000000000000..566a35916f33c285654e9257fed31d3edae82188 --- /dev/null +++ b/data/rewardbench/nvidia/Nemotron-4-340B-Reward/fd38add4-a301-4bae-a0f5-63a59e979d7a.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/nvidia_Nemotron-4-340B-Reward/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "nvidia/Nemotron-4-340B-Reward", + "id": "nvidia/Nemotron-4-340B-Reward", + "developer": "nvidia", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.92 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9581 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8706 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9149 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9363 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-3.5-turbo-0125/de0e882c-3cbc-44f0-8e2e-4712dc772602.json b/data/rewardbench/openai/gpt-3.5-turbo-0125/de0e882c-3cbc-44f0-8e2e-4712dc772602.json new file mode 100644 index 0000000000000000000000000000000000000000..591051b59f3ff89aba850884b8d3fa6741c9bf9c --- /dev/null +++ b/data/rewardbench/openai/gpt-3.5-turbo-0125/de0e882c-3cbc-44f0-8e2e-4712dc772602.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-3.5-turbo-0125/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-3.5-turbo-0125", + "id": "openai/gpt-3.5-turbo-0125", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6534 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4452 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6547 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5912 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6548 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4-0125-preview/cbefd48a-7036-4c1b-92d1-d785c6b3f808.json b/data/rewardbench/openai/gpt-4-0125-preview/cbefd48a-7036-4c1b-92d1-d785c6b3f808.json new file mode 100644 index 0000000000000000000000000000000000000000..48887ea92f73048b703e0cdee191a202000ce586 --- /dev/null +++ b/data/rewardbench/openai/gpt-4-0125-preview/cbefd48a-7036-4c1b-92d1-d785c6b3f808.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-4-0125-preview/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4-0125-preview", + "id": "openai/gpt-4-0125-preview", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8434 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7434 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8692 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7085 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4-turbo-2024-04-09/346fba29-4e65-4003-a849-2d5ee7b0a7a7.json b/data/rewardbench/openai/gpt-4-turbo-2024-04-09/346fba29-4e65-4003-a849-2d5ee7b0a7a7.json new file mode 100644 index 0000000000000000000000000000000000000000..8ee3b0bbf840331992793be4067159444666b414 --- /dev/null +++ b/data/rewardbench/openai/gpt-4-turbo-2024-04-09/346fba29-4e65-4003-a849-2d5ee7b0a7a7.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-4-turbo-2024-04-09/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4-turbo-2024-04-09", + "id": "openai/gpt-4-turbo-2024-04-09", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8395 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7544 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8757 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.827 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7363 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4.1-2025-04-14/dd2f9bdf-9f6f-4934-9c4c-85a131078412.json b/data/rewardbench/openai/gpt-4.1-2025-04-14/dd2f9bdf-9f6f-4934-9c4c-85a131078412.json new file mode 100644 index 0000000000000000000000000000000000000000..cb6ffddfd3b69dafea43a59c0b5fd2f0ad81fec3 --- /dev/null +++ b/data/rewardbench/openai/gpt-4.1-2025-04-14/dd2f9bdf-9f6f-4934-9c4c-85a131078412.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openai_gpt-4.1-2025-04-14/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4.1-2025-04-14", + "id": "openai/gpt-4.1-2025-04-14", + "developer": "openai", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7232 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8289 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3974 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6521 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8726 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7338 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8542 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4.1-mini-2025-04-14/4e48851c-05ab-42db-b141-fc2336f6d6ac.json b/data/rewardbench/openai/gpt-4.1-mini-2025-04-14/4e48851c-05ab-42db-b141-fc2336f6d6ac.json new file mode 100644 index 0000000000000000000000000000000000000000..adb0a213ad1c192f1f8fdf2e7c15c8a0c4971125 --- /dev/null +++ b/data/rewardbench/openai/gpt-4.1-mini-2025-04-14/4e48851c-05ab-42db-b141-fc2336f6d6ac.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openai_gpt-4.1-mini-2025-04-14/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4.1-mini-2025-04-14", + "id": "openai/gpt-4.1-mini-2025-04-14", + "developer": "openai", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6573 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6084 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4125 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7213 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7265 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7354 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.74 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4.1-nano-2025-04-14/da62a5e4-95ab-45e0-8e78-c9db94ca63ac.json b/data/rewardbench/openai/gpt-4.1-nano-2025-04-14/da62a5e4-95ab-45e0-8e78-c9db94ca63ac.json new file mode 100644 index 0000000000000000000000000000000000000000..5b8c1d53d31dcfa7c0a10e4da7727954eeb97217 --- /dev/null +++ b/data/rewardbench/openai/gpt-4.1-nano-2025-04-14/da62a5e4-95ab-45e0-8e78-c9db94ca63ac.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openai_gpt-4.1-nano-2025-04-14/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4.1-nano-2025-04-14", + "id": "openai/gpt-4.1-nano-2025-04-14", + "developer": "openai", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4849 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4646 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2578 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5041 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7156 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.466 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5015 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4o-2024-05-13/66575407-59ee-4eb3-9fbb-cb02706bb9cf.json b/data/rewardbench/openai/gpt-4o-2024-05-13/66575407-59ee-4eb3-9fbb-cb02706bb9cf.json new file mode 100644 index 0000000000000000000000000000000000000000..d2b08e1e70b503f4a000c8286ab2f7737b17d802 --- /dev/null +++ b/data/rewardbench/openai/gpt-4o-2024-05-13/66575407-59ee-4eb3-9fbb-cb02706bb9cf.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-4o-2024-05-13/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4o-2024-05-13", + "id": "openai/gpt-4o-2024-05-13", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8327 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7039 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8649 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8487 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7262 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4o-2024-08-06/45ecb6f2-042c-48ad-be13-34ea2582c809.json b/data/rewardbench/openai/gpt-4o-2024-08-06/45ecb6f2-042c-48ad-be13-34ea2582c809.json new file mode 100644 index 0000000000000000000000000000000000000000..6a6e4509602ed11b6687ea3bf926b3613f758f83 --- /dev/null +++ b/data/rewardbench/openai/gpt-4o-2024-08-06/45ecb6f2-042c-48ad-be13-34ea2582c809.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-4o-2024-08-06/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4o-2024-08-06", + "id": "openai/gpt-4o-2024-08-06", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8673 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9609 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.761 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8811 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8661 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4o-2024-08-06/c9f2c93b-2046-4223-8257-734500866f2d.json b/data/rewardbench/openai/gpt-4o-2024-08-06/c9f2c93b-2046-4223-8257-734500866f2d.json new file mode 100644 index 0000000000000000000000000000000000000000..bc9d31460783d218bbddc4d854a9bc89c7d7cccd --- /dev/null +++ b/data/rewardbench/openai/gpt-4o-2024-08-06/c9f2c93b-2046-4223-8257-734500866f2d.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openai_gpt-4o-2024-08-06/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4o-2024-08-06", + "id": "openai/gpt-4o-2024-08-06", + "developer": "openai", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6493 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5684 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.623 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8619 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7293 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7819 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4o-mini-2024-07-18/8fb0c78b-1141-4f68-9563-ac7f0dd15989.json b/data/rewardbench/openai/gpt-4o-mini-2024-07-18/8fb0c78b-1141-4f68-9563-ac7f0dd15989.json new file mode 100644 index 0000000000000000000000000000000000000000..855aafe9395ca493ad03c2d1145e4b533204662b --- /dev/null +++ b/data/rewardbench/openai/gpt-4o-mini-2024-07-18/8fb0c78b-1141-4f68-9563-ac7f0dd15989.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openai_gpt-4o-mini-2024-07-18/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4o-mini-2024-07-18", + "id": "openai/gpt-4o-mini-2024-07-18", + "developer": "openai", + "additional_details": { + "model_type": "Generative RM" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5796 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4105 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5191 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7414 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6962 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openai/gpt-4o-mini-2024-07-18/9e3febed-4033-4436-8d70-14048e928231.json b/data/rewardbench/openai/gpt-4o-mini-2024-07-18/9e3febed-4033-4436-8d70-14048e928231.json new file mode 100644 index 0000000000000000000000000000000000000000..e8cb061535cf5da57f07a322eb36fd3c6ef53b4a --- /dev/null +++ b/data/rewardbench/openai/gpt-4o-mini-2024-07-18/9e3febed-4033-4436-8d70-14048e928231.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openai_gpt-4o-mini-2024-07-18/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openai/gpt-4o-mini-2024-07-18", + "id": "openai/gpt-4o-mini-2024-07-18", + "developer": "openai", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8007 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9497 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6075 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8081 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8374 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/Eurus-7b-kto/fe187308-0168-46a8-a38d-f34282fb9d28.json b/data/rewardbench/openbmb/Eurus-7b-kto/fe187308-0168-46a8-a38d-f34282fb9d28.json new file mode 100644 index 0000000000000000000000000000000000000000..d5eac2038282e6bd88cc235dfa1ff285c1a5a0df --- /dev/null +++ b/data/rewardbench/openbmb/Eurus-7b-kto/fe187308-0168-46a8-a38d-f34282fb9d28.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openbmb_Eurus-7b-kto/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/Eurus-7b-kto", + "id": "openbmb/Eurus-7b-kto", + "developer": "openbmb", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.69 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9525 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5373 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6054 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7467 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5261 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/Eurus-RM-7b/5bb1f2d0-4674-4609-bddd-d2ed8b20fb59.json b/data/rewardbench/openbmb/Eurus-RM-7b/5bb1f2d0-4674-4609-bddd-d2ed8b20fb59.json new file mode 100644 index 0000000000000000000000000000000000000000..084575d0c0946497b349dde1f4b97a0a652c69f0 --- /dev/null +++ b/data/rewardbench/openbmb/Eurus-RM-7b/5bb1f2d0-4674-4609-bddd-d2ed8b20fb59.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openbmb_Eurus-RM-7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/Eurus-RM-7b", + "id": "openbmb/Eurus-RM-7b", + "developer": "openbmb", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5806 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5683 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6267 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7475 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5972 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/Eurus-RM-7b/c2e5a8ed-7212-490a-883b-6ce4fcdd4ad1.json b/data/rewardbench/openbmb/Eurus-RM-7b/c2e5a8ed-7212-490a-883b-6ce4fcdd4ad1.json new file mode 100644 index 0000000000000000000000000000000000000000..cedd88ae75d52f91b3c351e3942bbf21a12e4285 --- /dev/null +++ b/data/rewardbench/openbmb/Eurus-RM-7b/c2e5a8ed-7212-490a-883b-6ce4fcdd4ad1.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openbmb_Eurus-RM-7b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/Eurus-RM-7b", + "id": "openbmb/Eurus-RM-7b", + "developer": "openbmb", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8159 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9804 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6557 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8135 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8633 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7172 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/MiniCPM-2B-dpo-fp32/13c8397c-68c9-41bd-b0d6-ed75ac0825c7.json b/data/rewardbench/openbmb/MiniCPM-2B-dpo-fp32/13c8397c-68c9-41bd-b0d6-ed75ac0825c7.json new file mode 100644 index 0000000000000000000000000000000000000000..9f56eebd45de9e9441f04953bcf2fa5e5320b397 --- /dev/null +++ b/data/rewardbench/openbmb/MiniCPM-2B-dpo-fp32/13c8397c-68c9-41bd-b0d6-ed75ac0825c7.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openbmb_MiniCPM-2B-dpo-fp32/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/MiniCPM-2B-dpo-fp32", + "id": "openbmb/MiniCPM-2B-dpo-fp32", + "developer": "openbmb", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.673 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8911 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4934 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.573 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8233 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4958 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/UltraRM-13b/894f11bc-eb10-45e4-b364-8a105b9f3131.json b/data/rewardbench/openbmb/UltraRM-13b/894f11bc-eb10-45e4-b364-8a105b9f3131.json new file mode 100644 index 0000000000000000000000000000000000000000..749d6e31e2f8f1ff33e4ab14781a4e624b433d7b --- /dev/null +++ b/data/rewardbench/openbmb/UltraRM-13b/894f11bc-eb10-45e4-b364-8a105b9f3131.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/openbmb_UltraRM-13b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/UltraRM-13b", + "id": "openbmb/UltraRM-13b", + "developer": "openbmb", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6903 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5548 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5986 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6244 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7294 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/openbmb/UltraRM-13b/ee2653d2-79f5-42ed-bdc9-abfe430fc64c.json b/data/rewardbench/openbmb/UltraRM-13b/ee2653d2-79f5-42ed-bdc9-abfe430fc64c.json new file mode 100644 index 0000000000000000000000000000000000000000..de8cf52cec38799423b570c603905d7a19139728 --- /dev/null +++ b/data/rewardbench/openbmb/UltraRM-13b/ee2653d2-79f5-42ed-bdc9-abfe430fc64c.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/openbmb_UltraRM-13b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "openbmb/UltraRM-13b", + "id": "openbmb/UltraRM-13b", + "developer": "openbmb", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4683 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5063 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3312 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5519 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5089 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6081 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3036 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/opencompass/CompassJudger-1-1.5B-Instruct/6f131a84-b804-45be-8b4d-d74719d3e7da.json b/data/rewardbench/opencompass/CompassJudger-1-1.5B-Instruct/6f131a84-b804-45be-8b4d-d74719d3e7da.json new file mode 100644 index 0000000000000000000000000000000000000000..e655813a7bc0d670253b02de915429bef0174a55 --- /dev/null +++ b/data/rewardbench/opencompass/CompassJudger-1-1.5B-Instruct/6f131a84-b804-45be-8b4d-d74719d3e7da.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/opencompass_CompassJudger-1-1.5B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "opencompass/CompassJudger-1-1.5B-Instruct", + "id": "opencompass/CompassJudger-1-1.5B-Instruct", + "developer": "opencompass", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7344 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9637 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4923 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7818 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6999 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/opencompass/CompassJudger-1-14B-Instruct/c5adf60d-717a-45ed-8560-56580fad6091.json b/data/rewardbench/opencompass/CompassJudger-1-14B-Instruct/c5adf60d-717a-45ed-8560-56580fad6091.json new file mode 100644 index 0000000000000000000000000000000000000000..bf0ecb1bf739da92ee6aca3c14fd9389b6871c48 --- /dev/null +++ b/data/rewardbench/opencompass/CompassJudger-1-14B-Instruct/c5adf60d-717a-45ed-8560-56580fad6091.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/opencompass_CompassJudger-1-14B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "opencompass/CompassJudger-1-14B-Instruct", + "id": "opencompass/CompassJudger-1-14B-Instruct", + "developer": "opencompass", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8409 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9749 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6228 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8392 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9268 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/opencompass/CompassJudger-1-32B-Instruct/6d747303-83bc-4515-b202-462b1c8651e4.json b/data/rewardbench/opencompass/CompassJudger-1-32B-Instruct/6d747303-83bc-4515-b202-462b1c8651e4.json new file mode 100644 index 0000000000000000000000000000000000000000..c63d698e95bf19803fc3d9bfaffcba2635586571 --- /dev/null +++ b/data/rewardbench/opencompass/CompassJudger-1-32B-Instruct/6d747303-83bc-4515-b202-462b1c8651e4.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/opencompass_CompassJudger-1-32B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "opencompass/CompassJudger-1-32B-Instruct", + "id": "opencompass/CompassJudger-1-32B-Instruct", + "developer": "opencompass", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8522 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9804 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6513 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8527 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9244 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/opencompass/CompassJudger-1-7B-Instruct/08c8d9d7-fa54-45aa-84e6-ebd4b444b1e0.json b/data/rewardbench/opencompass/CompassJudger-1-7B-Instruct/08c8d9d7-fa54-45aa-84e6-ebd4b444b1e0.json new file mode 100644 index 0000000000000000000000000000000000000000..67de7fb3fee0863ea56a686d411a4a26628565a4 --- /dev/null +++ b/data/rewardbench/opencompass/CompassJudger-1-7B-Instruct/08c8d9d7-fa54-45aa-84e6-ebd4b444b1e0.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/opencompass_CompassJudger-1-7B-Instruct/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "opencompass/CompassJudger-1-7B-Instruct", + "id": "opencompass/CompassJudger-1-7B-Instruct", + "developer": "opencompass", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8317 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9777 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6096 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8446 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8948 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/prometheus-eval/prometheus-7b-v2.0/5b85d051-217b-4499-8f37-440dca222878.json b/data/rewardbench/prometheus-eval/prometheus-7b-v2.0/5b85d051-217b-4499-8f37-440dca222878.json new file mode 100644 index 0000000000000000000000000000000000000000..5185c04556eb9d382c490c2ca5c3973df43de65a --- /dev/null +++ b/data/rewardbench/prometheus-eval/prometheus-7b-v2.0/5b85d051-217b-4499-8f37-440dca222878.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/prometheus-eval_prometheus-7b-v2.0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "prometheus-eval/prometheus-7b-v2.0", + "id": "prometheus-eval/prometheus-7b-v2.0", + "developer": "prometheus-eval", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7204 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8547 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4912 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7709 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7648 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/prometheus-eval/prometheus-8x7b-v2.0/4dfc1d75-fdf9-475e-8800-ac63cda6901d.json b/data/rewardbench/prometheus-eval/prometheus-8x7b-v2.0/4dfc1d75-fdf9-475e-8800-ac63cda6901d.json new file mode 100644 index 0000000000000000000000000000000000000000..87be64c71972473222aee4edc0c5fe0c16610126 --- /dev/null +++ b/data/rewardbench/prometheus-eval/prometheus-8x7b-v2.0/4dfc1d75-fdf9-475e-8800-ac63cda6901d.json @@ -0,0 +1,116 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/prometheus-eval_prometheus-8x7b-v2.0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "prometheus-eval/prometheus-8x7b-v2.0", + "id": "prometheus-eval/prometheus-8x7b-v2.0", + "developer": "prometheus-eval", + "additional_details": { + "model_type": "Generative" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7451 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9302 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4715 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8047 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.774 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/4bcf0885-b8b8-4d06-9716-c938353c6d2d.json b/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/4bcf0885-b8b8-4d06-9716-c938353c6d2d.json new file mode 100644 index 0000000000000000000000000000000000000000..392ad373992be792ad7f58515bdc408c2d31da81 --- /dev/null +++ b/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/4bcf0885-b8b8-4d06-9716-c938353c6d2d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/sfairXC_FsfairX-LLaMA3-RM-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "sfairXC/FsfairX-LLaMA3-RM-v0.1", + "id": "sfairXC/FsfairX-LLaMA3-RM-v0.1", + "developer": "sfairXC", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8338 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9944 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6513 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8676 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8644 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7492 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/a2c84b3a-e539-41da-a6e2-ee6d7b766840.json b/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/a2c84b3a-e539-41da-a6e2-ee6d7b766840.json new file mode 100644 index 0000000000000000000000000000000000000000..f66bb009fe68f8d5bf386accf13dfb7c90b54a18 --- /dev/null +++ b/data/rewardbench/sfairXC/FsfairX-LLaMA3-RM-v0.1/a2c84b3a-e539-41da-a6e2-ee6d7b766840.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/sfairXC_FsfairX-LLaMA3-RM-v0.1/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "sfairXC/FsfairX-LLaMA3-RM-v0.1", + "id": "sfairXC/FsfairX-LLaMA3-RM-v0.1", + "developer": "sfairXC", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6292 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5916 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4188 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6284 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7667 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7051 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6647 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stabilityai/stable-code-instruct-3b/37b0a961-9122-4e77-ba85-78a7550d56b3.json b/data/rewardbench/stabilityai/stable-code-instruct-3b/37b0a961-9122-4e77-ba85-78a7550d56b3.json new file mode 100644 index 0000000000000000000000000000000000000000..134ec7d3e8cc9cf859cc439c1ac946f34991abc3 --- /dev/null +++ b/data/rewardbench/stabilityai/stable-code-instruct-3b/37b0a961-9122-4e77-ba85-78a7550d56b3.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stabilityai_stable-code-instruct-3b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stabilityai/stable-code-instruct-3b", + "id": "stabilityai/stable-code-instruct-3b", + "developer": "stabilityai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6216 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5782 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5855 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6554 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7528 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4506 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stabilityai/stablelm-2-12b-chat/e391ac88-3021-414b-8a4e-bf55e93ae8c9.json b/data/rewardbench/stabilityai/stablelm-2-12b-chat/e391ac88-3021-414b-8a4e-bf55e93ae8c9.json new file mode 100644 index 0000000000000000000000000000000000000000..dbba1bda1bbb70e2234382a73fb1a6d0a35e0252 --- /dev/null +++ b/data/rewardbench/stabilityai/stablelm-2-12b-chat/e391ac88-3021-414b-8a4e-bf55e93ae8c9.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stabilityai_stablelm-2-12b-chat/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stabilityai/stablelm-2-12b-chat", + "id": "stabilityai/stablelm-2-12b-chat", + "developer": "stabilityai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7642 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5548 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7811 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8945 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4839 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stabilityai/stablelm-2-zephyr-1_6b/e1146920-8bea-4939-9cfd-13b697087c0d.json b/data/rewardbench/stabilityai/stablelm-2-zephyr-1_6b/e1146920-8bea-4939-9cfd-13b697087c0d.json new file mode 100644 index 0000000000000000000000000000000000000000..b7af282815843b5fa92c76cda00b76d172164fdc --- /dev/null +++ b/data/rewardbench/stabilityai/stablelm-2-zephyr-1_6b/e1146920-8bea-4939-9cfd-13b697087c0d.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stabilityai_stablelm-2-zephyr-1_6b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stabilityai/stablelm-2-zephyr-1_6b", + "id": "stabilityai/stablelm-2-zephyr-1_6b", + "developer": "stabilityai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6574 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4671 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6027 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6784 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4868 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stabilityai/stablelm-zephyr-3b/2e68c89a-536d-4e2b-9421-5c30231fda6b.json b/data/rewardbench/stabilityai/stablelm-zephyr-3b/2e68c89a-536d-4e2b-9421-5c30231fda6b.json new file mode 100644 index 0000000000000000000000000000000000000000..9390c6c9a6869dec73ae8be32c6392c62a5228ed --- /dev/null +++ b/data/rewardbench/stabilityai/stablelm-zephyr-3b/2e68c89a-536d-4e2b-9421-5c30231fda6b.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stabilityai_stablelm-zephyr-3b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stabilityai/stablelm-zephyr-3b", + "id": "stabilityai/stablelm-zephyr-3b", + "developer": "stabilityai", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7146 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8631 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6009 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7405 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7573 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5075 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-large/fe677973-e5b3-419e-8efe-3ec2037b0346.json b/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-large/fe677973-e5b3-419e-8efe-3ec2037b0346.json new file mode 100644 index 0000000000000000000000000000000000000000..7515b0c233c863e195fafcded0886c5d340e4384 --- /dev/null +++ b/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-large/fe677973-e5b3-419e-8efe-3ec2037b0346.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stanfordnlp_SteamSHP-flan-t5-large/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stanfordnlp/SteamSHP-flan-t5-large", + "id": "stanfordnlp/SteamSHP-flan-t5-large", + "developer": "stanfordnlp", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4962 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8575 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3311 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3743 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3563 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6273 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-xl/e2111cd3-e584-4854-86dd-77226e11efa2.json b/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-xl/e2111cd3-e584-4854-86dd-77226e11efa2.json new file mode 100644 index 0000000000000000000000000000000000000000..ea221779e52cca313e83431430531b8ca1f2e450 --- /dev/null +++ b/data/rewardbench/stanfordnlp/SteamSHP-flan-t5-xl/e2111cd3-e584-4854-86dd-77226e11efa2.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/stanfordnlp_SteamSHP-flan-t5-xl/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "stanfordnlp/SteamSHP-flan-t5-xl", + "id": "stanfordnlp/SteamSHP-flan-t5-xl", + "developer": "stanfordnlp", + "additional_details": { + "model_type": "Custom Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5135 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8547 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3684 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3784 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3841 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6498 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/upstage/SOLAR-10.7B-Instruct-v1.0/7ea0ea12-2890-4024-92ce-50e193440c8b.json b/data/rewardbench/upstage/SOLAR-10.7B-Instruct-v1.0/7ea0ea12-2890-4024-92ce-50e193440c8b.json new file mode 100644 index 0000000000000000000000000000000000000000..08250f8d159586e0012d5999c4438a3423e12e8e --- /dev/null +++ b/data/rewardbench/upstage/SOLAR-10.7B-Instruct-v1.0/7ea0ea12-2890-4024-92ce-50e193440c8b.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/upstage_SOLAR-10.7B-Instruct-v1.0/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "upstage/SOLAR-10.7B-Instruct-v1.0", + "id": "upstage/SOLAR-10.7B-Instruct-v1.0", + "developer": "upstage", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7391 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8156 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6864 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8514 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7252 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4949 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/wenbopan/Faro-Yi-9B-DPO/b6df0a75-e362-4b7a-8ca2-0e09dbb8051a.json b/data/rewardbench/wenbopan/Faro-Yi-9B-DPO/b6df0a75-e362-4b7a-8ca2-0e09dbb8051a.json new file mode 100644 index 0000000000000000000000000000000000000000..b4ac331db96775d30194bb05bb56640eaf568451 --- /dev/null +++ b/data/rewardbench/wenbopan/Faro-Yi-9B-DPO/b6df0a75-e362-4b7a-8ca2-0e09dbb8051a.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/wenbopan_Faro-Yi-9B-DPO/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "wenbopan/Faro-Yi-9B-DPO", + "id": "wenbopan/Faro-Yi-9B-DPO", + "developer": "wenbopan", + "additional_details": { + "model_type": "DPO" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6461 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9218 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5307 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5514 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5839 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6395 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Gemma-2B/0f73cf10-fc76-4d18-8ba6-0b044aa1ff4a.json b/data/rewardbench/weqweasdas/RM-Gemma-2B/0f73cf10-fc76-4d18-8ba6-0b044aa1ff4a.json new file mode 100644 index 0000000000000000000000000000000000000000..2bed332ef27978e5067834d282219d7cc1850811 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Gemma-2B/0f73cf10-fc76-4d18-8ba6-0b044aa1ff4a.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/weqweasdas_RM-Gemma-2B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Gemma-2B", + "id": "weqweasdas/RM-Gemma-2B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3057 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3705 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2812 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4317 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3311 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2343 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.1851 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Gemma-2B/e3821ed1-f207-4b75-95c0-64704c59ecc0.json b/data/rewardbench/weqweasdas/RM-Gemma-2B/e3821ed1-f207-4b75-95c0-64704c59ecc0.json new file mode 100644 index 0000000000000000000000000000000000000000..20e5f6ec642f705aa6199644da26fb4d397dd4ce --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Gemma-2B/e3821ed1-f207-4b75-95c0-64704c59ecc0.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/weqweasdas_RM-Gemma-2B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Gemma-2B", + "id": "weqweasdas/RM-Gemma-2B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6549 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9441 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4079 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4986 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7637 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6652 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Gemma-7B-4096/d8cd4c16-c168-4835-82f6-a7419cb48318.json b/data/rewardbench/weqweasdas/RM-Gemma-7B-4096/d8cd4c16-c168-4835-82f6-a7419cb48318.json new file mode 100644 index 0000000000000000000000000000000000000000..be2240fb9f62d56eaea1a668840487489c469333 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Gemma-7B-4096/d8cd4c16-c168-4835-82f6-a7419cb48318.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/weqweasdas_RM-Gemma-7B-4096/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Gemma-7B-4096", + "id": "weqweasdas/RM-Gemma-7B-4096", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6922 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9497 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5022 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5608 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7511 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7024 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Gemma-7B/0e1d7f84-aee1-4f07-bcc9-2ca817962034.json b/data/rewardbench/weqweasdas/RM-Gemma-7B/0e1d7f84-aee1-4f07-bcc9-2ca817962034.json new file mode 100644 index 0000000000000000000000000000000000000000..60a534f2efb1c7fa35dcd4cc82b2c7b066c1be53 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Gemma-7B/0e1d7f84-aee1-4f07-bcc9-2ca817962034.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/weqweasdas_RM-Gemma-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Gemma-7B", + "id": "weqweasdas/RM-Gemma-7B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4826 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4926 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3937 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6066 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4822 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.497 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4232 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Gemma-7B/a2215713-f5aa-41ec-8608-534d9174e892.json b/data/rewardbench/weqweasdas/RM-Gemma-7B/a2215713-f5aa-41ec-8608-534d9174e892.json new file mode 100644 index 0000000000000000000000000000000000000000..b7b4964692703977b77a2e588e48f66f8c94dae8 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Gemma-7B/a2215713-f5aa-41ec-8608-534d9174e892.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/weqweasdas_RM-Gemma-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Gemma-7B", + "id": "weqweasdas/RM-Gemma-7B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6967 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9693 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4978 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5784 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7362 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7069 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Mistral-7B/374cee3a-57fa-43fb-a4e9-432ff006b7c7.json b/data/rewardbench/weqweasdas/RM-Mistral-7B/374cee3a-57fa-43fb-a4e9-432ff006b7c7.json new file mode 100644 index 0000000000000000000000000000000000000000..ec430dbdace78dce3d77211c7b4282be8175e471 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Mistral-7B/374cee3a-57fa-43fb-a4e9-432ff006b7c7.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/weqweasdas_RM-Mistral-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Mistral-7B", + "id": "weqweasdas/RM-Mistral-7B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.596 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5937 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3438 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5956 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6911 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7293 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6226 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/RM-Mistral-7B/5627d18f-ee8a-4724-bd0e-29610f07d834.json b/data/rewardbench/weqweasdas/RM-Mistral-7B/5627d18f-ee8a-4724-bd0e-29610f07d834.json new file mode 100644 index 0000000000000000000000000000000000000000..0a14a521da2f7fabf63166b720930d21ce43e3c1 --- /dev/null +++ b/data/rewardbench/weqweasdas/RM-Mistral-7B/5627d18f-ee8a-4724-bd0e-29610f07d834.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/weqweasdas_RM-Mistral-7B/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/RM-Mistral-7B", + "id": "weqweasdas/RM-Mistral-7B", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7982 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.9665 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6053 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8703 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.7736 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.753 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/95d84b17-20a7-436a-932b-5bd7bd1839af.json b/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/95d84b17-20a7-436a-932b-5bd7bd1839af.json new file mode 100644 index 0000000000000000000000000000000000000000..994463bc452d8aa41b3cd66eb1be6f1e4faf68fe --- /dev/null +++ b/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/95d84b17-20a7-436a-932b-5bd7bd1839af.json @@ -0,0 +1,152 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench-2/weqweasdas_hh_rlhf_rm_open_llama_3b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench 2", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/hh_rlhf_rm_open_llama_3b", + "id": "weqweasdas/hh_rlhf_rm_open_llama_3b", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench 2 Score (mean of all metrics)", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2498 + } + }, + { + "evaluation_name": "Factuality", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Factuality score - measures factual accuracy", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3642 + } + }, + { + "evaluation_name": "Precise IF", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Precise Instruction Following score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.275 + } + }, + { + "evaluation_name": "Math", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Math score - measures mathematical reasoning", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3497 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Safety score - measures safety awareness", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.24 + } + }, + { + "evaluation_name": "Focus", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Focus score - measures response focus", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.2384 + } + }, + { + "evaluation_name": "Ties", + "source_data": { + "dataset_name": "RewardBench 2", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench-2-results" + }, + "metric_config": { + "evaluation_description": "Ties score - ability to identify tie cases", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.0315 + } + } + ] +} \ No newline at end of file diff --git a/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/f8371a52-e5e9-4fa3-ba7d-6fc900966b04.json b/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/f8371a52-e5e9-4fa3-ba7d-6fc900966b04.json new file mode 100644 index 0000000000000000000000000000000000000000..838ae2e11f9972aba9f32344d68d2c1496b0ee76 --- /dev/null +++ b/data/rewardbench/weqweasdas/hh_rlhf_rm_open_llama_3b/f8371a52-e5e9-4fa3-ba7d-6fc900966b04.json @@ -0,0 +1,134 @@ +{ + "schema_version": "0.2.1", + "evaluation_id": "reward-bench/weqweasdas_hh_rlhf_rm_open_llama_3b/1783820827.6488361", + "retrieved_timestamp": "1783820827.6488361", + "source_metadata": { + "source_name": "RewardBench", + "source_type": "documentation", + "source_organization_name": "Allen Institute for AI", + "source_organization_url": "https://allenai.org", + "evaluator_relationship": "third_party" + }, + "eval_library": { + "name": "unknown", + "version": "unknown" + }, + "model_info": { + "name": "weqweasdas/hh_rlhf_rm_open_llama_3b", + "id": "weqweasdas/hh_rlhf_rm_open_llama_3b", + "developer": "weqweasdas", + "additional_details": { + "model_type": "Seq. Classifier" + } + }, + "evaluation_results": [ + { + "evaluation_name": "Score", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Overall RewardBench Score", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.5027 + } + }, + { + "evaluation_name": "Chat", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat accuracy - includes easy chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.8184 + } + }, + { + "evaluation_name": "Chat Hard", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Chat Hard accuracy - includes hard chat subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3728 + } + }, + { + "evaluation_name": "Safety", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Safety accuracy - includes safety subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.4149 + } + }, + { + "evaluation_name": "Reasoning", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Reasoning accuracy - includes code and math subsets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.3281 + } + }, + { + "evaluation_name": "Prior Sets (0.5 weight)", + "source_data": { + "dataset_name": "RewardBench", + "source_type": "hf_dataset", + "hf_repo": "allenai/reward-bench" + }, + "metric_config": { + "evaluation_description": "Prior Sets score (weighted 0.5) - includes test sets", + "lower_is_better": false, + "score_type": "continuous", + "min_score": 0.0, + "max_score": 1.0 + }, + "score_details": { + "score": 0.6564 + } + } + ] +} \ No newline at end of file diff --git a/data/run_report.json b/data/run_report.json index 1903d2601149c648498f1b64b5612c75dcf4b8a3..8c7b859f0d8066634602683192b5f2987a4991fe 100644 --- a/data/run_report.json +++ b/data/run_report.json @@ -1,48 +1,83 @@ { "adapters": { - "cocoabench": { - "error": " [--benchmark-version BENCHMARK_VERSION]\n [--eval-library-version EVAL_LIBRARY_VERSION]\n [--evaluation-timestamp EVALUATION_TIMESTAMP]\n [--public-source-url PUBLIC_SOURCE_URLS]\n [--benchmark-reference-url BENCHMARK_REFERENCE_URLS]\npython3 -m utils.cocoabench.adapter: error: the following arguments are required: --csv\n", - "execution_failed": true - }, - "hal": { - "data_changed": false, - "errors": [], - "execution_failed": false, - "failed_files": 0, - "valid_files": 246 - }, - "llm_stats": { + "global-mmlu-lite": { "data_changed": true, "errors": [], "execution_failed": false, "failed_files": 0, - "valid_files": 324 + "valid_files": 28 }, - "mt_bench": { + "hle": { "data_changed": false, "errors": [], "execution_failed": false, "failed_files": 0, - "valid_files": 34 + "valid_files": 50 }, - "swe_polybench": { - "error": "\n File \"/home/runner/work/every_eval_ever-cron/every_eval_ever-cron/.venv/lib/python3.14/site-packages/dill/_dill.py\", line 1217, in save_module_dict\n StockPickler.save_dict(pickler, obj)\n ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^\n File \"/opt/hostedtoolcache/Python/3.14.6/x64/lib/python3.14/pickle.py\", line 1058, in save_dict\n self._batch_setitems(obj.items(), obj)\n ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^\nTypeError: Pickler._batch_setitems() takes 2 positional arguments but 3 were given\n", + "livecodebenchpro": { + "error": "e \"\", line 88, in _run_code\n File \"/home/runner/work/every_eval_ever-cron/every_eval_ever-cron/utils/livecodebenchpro/adapter.py\", line 94, in \n main()\n ~~~~^^\n File \"/home/runner/work/every_eval_ever-cron/every_eval_ever-cron/utils/livecodebenchpro/adapter.py\", line 82, in main\n raise FileNotFoundError(f'No JSON files found in {DATA_DIR}')\nFileNotFoundError: No JSON files found in /home/runner/work/every_eval_ever-cron/every_eval_ever-cron/data/livecodebenchpro\n", "execution_failed": true }, - "terminal_bench_2": { + "mercor_eval": { + "error": "base_url}/benchmarks',\n headers=headers,\n timeout=DEFAULT_TIMEOUT,\n )\n File \"/home/runner/work/every_eval_ever-cron/every_eval_ever-cron/utils/mercor_eval/adapter.py\", line 217, in request_json\n raise RuntimeError(\n f'Failed to fetch Mercor endpoint {url}: {exc}'\n ) from exc\nRuntimeError: Failed to fetch Mercor endpoint https://coil.mercor.com/external/evals/v1/benchmarks: 401 Client Error: Unauthorized for url: https://coil.mercor.com/external/evals/v1/benchmarks\n", + "execution_failed": true + }, + "multi_swe_bench": { "data_changed": false, "errors": [], "execution_failed": false, "failed_files": 0, - "valid_files": 115 + "valid_files": 262 }, - "vals_ai": { - "data_changed": true, + "openeval": { + "data_changed": false, "errors": [], "execution_failed": false, "failed_files": 0, - "valid_files": 2118 + "valid_files": 1112 + }, + "rewardbench": { + "error": "Validation: 384 valid, 6 failed", + "errors": [ + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/llama-2-chat-7b-nectar-3.8m.json'", + "type": "io_error" + }, + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/llama-2-chat-nectar-180k.json'", + "type": "io_error" + }, + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized-700k.json'", + "type": "io_error" + }, + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/tulu-2-7b-rm-v0-nectar-binarized.json'", + "type": "io_error" + }, + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/tulu-2-7b-rm-v0.json'", + "type": "io_error" + }, + { + "loc": "(file)", + "msg": "[Errno 21] Is a directory: 'data/rewardbench/ai2/llama-2-chat-ultrafeedback-60k.jsonl'", + "type": "io_error" + } + ], + "execution_failed": false, + "failed_files": 6, + "valid_files": 384 + }, + "swe_bench_verified": { + "error": "\n File \"/home/runner/work/every_eval_ever-cron/every_eval_ever-cron/.venv/lib/python3.14/site-packages/dill/_dill.py\", line 1217, in save_module_dict\n StockPickler.save_dict(pickler, obj)\n ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^\n File \"/opt/hostedtoolcache/Python/3.14.6/x64/lib/python3.14/pickle.py\", line 1058, in save_dict\n self._batch_setitems(obj.items(), obj)\n ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^\nTypeError: Pickler._batch_setitems() takes 2 positional arguments but 3 were given\n", + "execution_failed": true } }, - "date": "2026-07-11" + "date": "2026-07-12" } \ No newline at end of file