Scheduled Commit

data/retrieval_individual-cc11c9ee-fc93-4cd6-a4b5-246a7f21038c.jsonl

@@ -96,3 +96,7 @@
 {"tstamp": 1722332320.4506, "task_type": "retrieval", "type": "chat", "model": "mixedbread-ai/mxbai-embed-large-v1", "gen_params": {}, "start": 1722332318.2864, "finish": 1722332320.4506, "ip": "", "conv_id": "09f9e3088f8f45be984082f34dd57d21", "model_name": "mixedbread-ai/mxbai-embed-large-v1", "prompt": "Why can't my lawn get too tall in the suburbs?", "output": [["Why can't my lawn get too tall in the suburbs?", "Lawn\n\nReplacing turf grass with low-maintenance groundcovers or employing a variety of low-maintenance perennials, trees and shrubs can be a good alternative to traditional lawn spaces, especially in hard-to-grow or hard-to-mow areas, as it can reduce maintenance requirements, associated pollution and offers higher aesthetic and wildlife value. Growing a mixed variety of flowering plants instead of turfgrass is sometimes referred to as meadowscaping.\nNon-productive space\nLawns take up space that could otherwise be used more productively, such as for urban agriculture or home gardening. This is the case in many cities and suburbs in the United States, where open or unused spaces are \"not generally a result of a positive decision to leave room for some use, but rather is an expression of a pastoral aesthetic norm that prizes spacious lawns and the zoning restrictions and neighborhood covenants that give these norms the force of law.\"\nIn urban and suburban spaces, growing food in front yards and parking strips can not only provide fresh produce but also be a source of neighborhood pride. While converting lawn space into strictly utilitarian farms is not common, incorporating edible plants into front yards with sustainable and aesthetically pleasing design is of growing interest in the United States."]], "corpus": "wikipedia"}
 {"tstamp": 1722342704.0839, "task_type": "retrieval", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722342701.7306, "finish": 1722342704.0839, "ip": "", "conv_id": "c1e4e2dd9dac48b8a6c28f18bf8f6a8d", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": "who is the actor that plays ducky on ncis", "output": [["who is the actor that plays ducky on ncis", "List of NCIS cast members\n\nBelow is a list of actors and actresses that are part of the cast of the American police procedural drama television series NCIS.\nCast members\nThis table lists characters who have appeared in two or more episodes (excluding flashbacks) and the cast members who have portrayed them.\nColor key:\n† Indicates the actor/actress is deceased while the series is in production\nCrossovers\nJAG\nMark Harmon, Michael Weatherly, David McCallum, and Pauley Perrette all appeared in two episodes of season eight of JAG that served as a back door pilot.\nPatrick Labyorteaux appeared as his JAG character in three episodes of NCIS (one each in seasons 1, 14, and 15)\nJohn M. Jackson appeared as his JAG character in one episode of season ten of NCIS.\nAdam Baldwin appeared as his JAG character in one episode of season one of NCIS.\nAlicia Coppola appeared as her JAG character in three episodes of NCIS (one in season one and two in season two)."]], "corpus": "wikipedia"}
 {"tstamp": 1722342704.0839, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722342701.7306, "finish": 1722342704.0839, "ip": "", "conv_id": "f5f32f5055aa4f4f9d7c2e50fc3cba50", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "who is the actor that plays ducky on ncis", "output": [["who is the actor that plays ducky on ncis", "David McCallum\n\nDavid Keith McCallum (19 September 1933 – 25 September 2023) was a Scottish actor and musician, based in the United States. He gained wide recognition in the 1960s for playing secret agent Illya Kuryakin in the television series The Man from U.N.C.L.E (1964–1968). His other notable television roles include Simon Carter in Colditz (1972–1974) and Steel in Sapphire & Steel (1979–1982). Beginning in 2003, McCallum gained renewed international popularity for his role as NCIS medical examiner Dr. Donald \"Ducky\" Mallard in the American CBS television series NCIS, which he played for 20 seasons until his death in 2023. In film roles, McCallum notably appeared in The Great Escape (1963), and as Judas Iscariot in The Greatest Story Ever Told (1965).\nEarly life\nDavid Keith McCallum was born on 19 September 1933 in Glasgow the second of two sons of orchestral violinist David Fotheringham and Dorothy McCallum (née Dorman), a cellist. When he was three, his family moved to London for his father to play as the leader of the London Philharmonic Orchestra. Early in the Second World War, he was evacuated back to Scotland, where he lived with his mother at Gartocharn by Loch Lomond."]], "corpus": "wikipedia"}
+{"tstamp": 1722355756.6623, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722355754.3574, "finish": 1722355756.6623, "ip": "", "conv_id": "5f22cd5fed984a3094edf1dc060a61ff", "model_name": "GritLM/GritLM-7B", "prompt": "Which test was devised to determine whether robots can think?", "output": [["Which test was devised to determine whether robots can think?", "Turing test\n\nThe test was introduced by Turing in his 1950 paper \"Computing Machinery and Intelligence\" while working at the University of Manchester. It opens with the words: \"I propose to consider the question, 'Can machines think? Because \"thinking\" is difficult to define, Turing chooses to \"replace the question by another, which is closely related to it and is expressed in relatively unambiguous words.\" Turing describes the new form of the problem in terms of a three-person game called the \"imitation game\", in which an interrogator asks questions of a man and a woman in another room in order to determine the correct sex of the two players. Turing's new question is: \"Are there imaginable digital computers which would do well in the imitation game?\" This question, Turing believed, was one that could actually be answered. In the remainder of the paper, he argued against all the major objections to the proposition that \"machines can think\".\nSince Turing introduced his test, it has been both highly influential and widely criticized, and has become an important concept in the philosophy of artificial intelligence. Philosopher John Searle would comment on the Turing test in his Chinese room argument, a thought experiment that stipulates that a machine cannot have a \"mind\", \"understanding\", or \"consciousness\", regardless of how intelligently or human-like the program may make the computer behave. Searle criticizes Turing's test and claims it is insufficient to detect the presence of consciousness.\nHistory"]], "corpus": "wikipedia"}
+{"tstamp": 1722355756.6623, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722355754.3574, "finish": 1722355756.6623, "ip": "", "conv_id": "9ef794b8e63b463cb3b5ed634b80ed97", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "Which test was devised to determine whether robots can think?", "output": [["Which test was devised to determine whether robots can think?", "Turing test\n\nThe Turing test, originally called the imitation game by Alan Turing in 1950, is a test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human. Turing proposed that a human evaluator would judge natural language conversations between a human and a machine designed to generate human-like responses. The evaluator would be aware that one of the two partners in conversation was a machine, and all participants would be separated from one another. The conversation would be limited to a text-only channel, such as a computer keyboard and screen, so the result would not depend on the machine's ability to render words as speech. If the evaluator could not reliably tell the machine from the human, the machine would be said to have passed the test. The test results would not depend on the machine's ability to give correct answers to questions, only on how closely its answers resembled those a human would give. Since the Turing test is a test of indistinguishability in performance capacity, the verbal version generalizes naturally to all of human performance capacity, verbal as well as nonverbal (robotic)."]], "corpus": "wikipedia"}
+{"tstamp": 1722355846.2604, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722355846.0308, "finish": 1722355846.2604, "ip": "", "conv_id": "cf328bd4ce6842b3b75994dc47cf36e3", "model_name": "GritLM/GritLM-7B", "prompt": "I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "output": [["I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "Title: MTEB: Massive Text Embedding Benchmark\n\nAbstract: Text embeddings are commonly evaluated on a small set of datasets from a single task not covering their possible applications to other tasks. It is unclear whether state-of-the-art embeddings on semantic textual similarity (STS) can be equally well applied to other tasks like clustering or reranking. This makes progress in the field difficult to track, as various models are constantly being proposed without proper evaluation. To solve this problem, we introduce the Massive Text Embedding Benchmark (MTEB). MTEB spans 8 embedding tasks covering a total of 58 datasets and 112 languages. Through the benchmarking of 33 models on MTEB, we establish the most comprehensive benchmark of text embeddings to date. We find that no particular text embedding method dominates across all tasks. This suggests that the field has yet to converge on a universal text embedding method and scale it up sufficiently to provide state-of-the-art results on all embedding tasks. MTEB comes with open-source code and a public leaderboard at https://github.com/embeddings-benchmark/mteb."]], "corpus": "arxiv"}
+{"tstamp": 1722355846.2604, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722355846.0308, "finish": 1722355846.2604, "ip": "", "conv_id": "e612be7a0fc542c981bdede00d9d4ced", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "output": [["I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "Title: MTEB: Massive Text Embedding Benchmark\n\nAbstract: Text embeddings are commonly evaluated on a small set of datasets from a single task not covering their possible applications to other tasks. It is unclear whether state-of-the-art embeddings on semantic textual similarity (STS) can be equally well applied to other tasks like clustering or reranking. This makes progress in the field difficult to track, as various models are constantly being proposed without proper evaluation. To solve this problem, we introduce the Massive Text Embedding Benchmark (MTEB). MTEB spans 8 embedding tasks covering a total of 58 datasets and 112 languages. Through the benchmarking of 33 models on MTEB, we establish the most comprehensive benchmark of text embeddings to date. We find that no particular text embedding method dominates across all tasks. This suggests that the field has yet to converge on a universal text embedding method and scale it up sufficiently to provide state-of-the-art results on all embedding tasks. MTEB comes with open-source code and a public leaderboard at https://github.com/embeddings-benchmark/mteb."]], "corpus": "arxiv"}

data/retrieval_side_by_side-cc11c9ee-fc93-4cd6-a4b5-246a7f21038c.jsonl

@@ -1 +1,3 @@
 {"tstamp": 1722332338.1269, "task_type": "retrieval", "type": "bothbadvote", "models": ["jinaai/jina-embeddings-v2-base-en", "mixedbread-ai/mxbai-embed-large-v1"], "ip": "", "0_conv_id": "804ddfe21ba240c19d8b2b8247e7fe5b", "0_model_name": "jinaai/jina-embeddings-v2-base-en", "0_prompt": "Why can't my lawn get too tall in the suburbs?", "0_output": [["Why can't my lawn get too tall in the suburbs?", "The Suburbs\n\n#2 – Billboards Top 10 Albums of 2010\n#2 – Magnets Top 20 Albums of 2010\n#2 – NMEs Top 75 Albums of 2010\n#2 – Relevant Magazines Top 10 Albums of 2010\n#2 – Stereogum's Top 50 Albums of 2010\n#2 – Times Top 10 Albums of 2010\n#2 – Under the Radars Top 50 Albums of 2010\n#3 – Spins 40 Best Albums of 2010\n#4 – MTV's 20 Best Albums of 2010\n#4 – Rolling Stones 30 Best Albums of 2010\n#7 – Paste Magazine's 50 Best Albums of 2010\n#9 – American Songwriters Top 50 Albums of 2010\n#11 – Drowned in Sound's Albums of the Year\n#11 – Pitchfork's Top 50 Albums of 2010\n#21 – Rough Trade Shops's Albums of the Year (UK)\n#23 – Robert Christgau's 2010 Dean's List\nGlide Magazines Top 20 Albums of 2010\nNPR's 50 Favorite Albums of 2010\nThe single “Ready to Start” was nominated for the Grammy Award for Best Rock Performance by a Duo or Group with Vocal.\nTrack listingNotes\"Suburban War\" is the second to last track on the vinyl version of the album.\n\"We Used to Wait\" finishes on an infinite loop at the end of Side 3 of the vinyl version."]], "0_corpus": "wikipedia", "1_conv_id": "09f9e3088f8f45be984082f34dd57d21", "1_model_name": "mixedbread-ai/mxbai-embed-large-v1", "1_prompt": "Why can't my lawn get too tall in the suburbs?", "1_output": [["Why can't my lawn get too tall in the suburbs?", "Lawn\n\nReplacing turf grass with low-maintenance groundcovers or employing a variety of low-maintenance perennials, trees and shrubs can be a good alternative to traditional lawn spaces, especially in hard-to-grow or hard-to-mow areas, as it can reduce maintenance requirements, associated pollution and offers higher aesthetic and wildlife value. Growing a mixed variety of flowering plants instead of turfgrass is sometimes referred to as meadowscaping.\nNon-productive space\nLawns take up space that could otherwise be used more productively, such as for urban agriculture or home gardening. This is the case in many cities and suburbs in the United States, where open or unused spaces are \"not generally a result of a positive decision to leave room for some use, but rather is an expression of a pastoral aesthetic norm that prizes spacious lawns and the zoning restrictions and neighborhood covenants that give these norms the force of law.\"\nIn urban and suburban spaces, growing food in front yards and parking strips can not only provide fresh produce but also be a source of neighborhood pride. While converting lawn space into strictly utilitarian farms is not common, incorporating edible plants into front yards with sustainable and aesthetically pleasing design is of growing interest in the United States."]], "1_corpus": "wikipedia"}
+{"tstamp": 1722355808.1071, "task_type": "retrieval", "type": "rightvote", "models": ["GritLM/GritLM-7B", "BAAI/bge-large-en-v1.5"], "ip": "", "0_conv_id": "5f22cd5fed984a3094edf1dc060a61ff", "0_model_name": "GritLM/GritLM-7B", "0_prompt": "Which test was devised to determine whether robots can think?", "0_output": [["Which test was devised to determine whether robots can think?", "Turing test\n\nThe test was introduced by Turing in his 1950 paper \"Computing Machinery and Intelligence\" while working at the University of Manchester. It opens with the words: \"I propose to consider the question, 'Can machines think? Because \"thinking\" is difficult to define, Turing chooses to \"replace the question by another, which is closely related to it and is expressed in relatively unambiguous words.\" Turing describes the new form of the problem in terms of a three-person game called the \"imitation game\", in which an interrogator asks questions of a man and a woman in another room in order to determine the correct sex of the two players. Turing's new question is: \"Are there imaginable digital computers which would do well in the imitation game?\" This question, Turing believed, was one that could actually be answered. In the remainder of the paper, he argued against all the major objections to the proposition that \"machines can think\".\nSince Turing introduced his test, it has been both highly influential and widely criticized, and has become an important concept in the philosophy of artificial intelligence. Philosopher John Searle would comment on the Turing test in his Chinese room argument, a thought experiment that stipulates that a machine cannot have a \"mind\", \"understanding\", or \"consciousness\", regardless of how intelligently or human-like the program may make the computer behave. Searle criticizes Turing's test and claims it is insufficient to detect the presence of consciousness.\nHistory"]], "0_corpus": "wikipedia", "1_conv_id": "9ef794b8e63b463cb3b5ed634b80ed97", "1_model_name": "BAAI/bge-large-en-v1.5", "1_prompt": "Which test was devised to determine whether robots can think?", "1_output": [["Which test was devised to determine whether robots can think?", "Turing test\n\nThe Turing test, originally called the imitation game by Alan Turing in 1950, is a test of a machine's ability to exhibit intelligent behaviour equivalent to, or indistinguishable from, that of a human. Turing proposed that a human evaluator would judge natural language conversations between a human and a machine designed to generate human-like responses. The evaluator would be aware that one of the two partners in conversation was a machine, and all participants would be separated from one another. The conversation would be limited to a text-only channel, such as a computer keyboard and screen, so the result would not depend on the machine's ability to render words as speech. If the evaluator could not reliably tell the machine from the human, the machine would be said to have passed the test. The test results would not depend on the machine's ability to give correct answers to questions, only on how closely its answers resembled those a human would give. Since the Turing test is a test of indistinguishability in performance capacity, the verbal version generalizes naturally to all of human performance capacity, verbal as well as nonverbal (robotic)."]], "1_corpus": "wikipedia"}
+{"tstamp": 1722355864.8668, "task_type": "retrieval", "type": "tievote", "models": ["GritLM/GritLM-7B", "BAAI/bge-large-en-v1.5"], "ip": "", "0_conv_id": "cf328bd4ce6842b3b75994dc47cf36e3", "0_model_name": "GritLM/GritLM-7B", "0_prompt": "I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "0_output": [["I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "Title: MTEB: Massive Text Embedding Benchmark\n\nAbstract: Text embeddings are commonly evaluated on a small set of datasets from a single task not covering their possible applications to other tasks. It is unclear whether state-of-the-art embeddings on semantic textual similarity (STS) can be equally well applied to other tasks like clustering or reranking. This makes progress in the field difficult to track, as various models are constantly being proposed without proper evaluation. To solve this problem, we introduce the Massive Text Embedding Benchmark (MTEB). MTEB spans 8 embedding tasks covering a total of 58 datasets and 112 languages. Through the benchmarking of 33 models on MTEB, we establish the most comprehensive benchmark of text embeddings to date. We find that no particular text embedding method dominates across all tasks. This suggests that the field has yet to converge on a universal text embedding method and scale it up sufficiently to provide state-of-the-art results on all embedding tasks. MTEB comes with open-source code and a public leaderboard at https://github.com/embeddings-benchmark/mteb."]], "0_corpus": "arxiv", "1_conv_id": "e612be7a0fc542c981bdede00d9d4ced", "1_model_name": "BAAI/bge-large-en-v1.5", "1_prompt": "I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "1_output": [["I am searching for a good and large-scale benchmark for testing the performance of text embeddings.", "Title: MTEB: Massive Text Embedding Benchmark\n\nAbstract: Text embeddings are commonly evaluated on a small set of datasets from a single task not covering their possible applications to other tasks. It is unclear whether state-of-the-art embeddings on semantic textual similarity (STS) can be equally well applied to other tasks like clustering or reranking. This makes progress in the field difficult to track, as various models are constantly being proposed without proper evaluation. To solve this problem, we introduce the Massive Text Embedding Benchmark (MTEB). MTEB spans 8 embedding tasks covering a total of 58 datasets and 112 languages. Through the benchmarking of 33 models on MTEB, we establish the most comprehensive benchmark of text embeddings to date. We find that no particular text embedding method dominates across all tasks. This suggests that the field has yet to converge on a universal text embedding method and scale it up sufficiently to provide state-of-the-art results on all embedding tasks. MTEB comes with open-source code and a public leaderboard at https://github.com/embeddings-benchmark/mteb."]], "1_corpus": "arxiv"}