Muennighoff
commited on
Scheduled Commit
Browse files
data/retrieval_individual-b9229914-47bc-4da8-a21b-89329fff8207.jsonl
CHANGED
@@ -23,3 +23,10 @@
|
|
23 |
{"tstamp": 1722306330.1591, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722306329.8317, "finish": 1722306330.1591, "ip": "", "conv_id": "940948f56e9c4f5297cb17f69935bd8b", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "when did scotland last qualify for world cup", "output": [["when did scotland last qualify for world cup", "Scotland at the FIFA World Cup\n\nThe World Cup consists of two parts, the qualification phase and the final phase (officially called the World Cup Finals). The qualification phase, which currently take place over the three years preceding the Finals, is used to determine which teams qualify for the Finals. The current format of the Finals involves 32 teams competing at venues within the host nation (or nations) over a period of about a month. The World Cup Finals is the most widely viewed sporting event in the world, with an estimated 715.1 million people watching the 2006 tournament final.\nHistory\n1930s\nScotland did not compete in the first three World Cup competitions, in 1930, 1934 and 1938. Because of a dispute with FIFA over \"broken-time\" payments to players, the SFA, with the football associations of England, Ireland and Wales, withdrew from FIFA in 1928, and did not rejoin as a permanent member until 1946.\n1950s\nThe readmission of the Scottish Football Association to FIFA in 1946 meant that Scotland were now eligible to enter the 1950 FIFA World Cup. FIFA advised that places would be awarded to the top two teams in the 1949–50 British Home Championship, but the SFA announced that Scotland would attend the finals only if Scotland won the competition. Scotland won their first two matches, but a 1–0 home defeat by England meant that the Scots finished as runners-up. The SFA stood by this proclamation, despite pleas to the contrary by the Scotland players, supported by England captain Billy Wright and the other England players. Sportswriter Brian Glanville later described the SFA as showing \"baffling insularity and pique\" in their refusal to send a team to the World Cup."]], "corpus": "wikipedia"}
|
24 |
{"tstamp": 1722306358.068, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722306357.4875, "finish": 1722306358.068, "ip": "", "conv_id": "be3fc968a3aa487ba8db65457d91fec3", "model_name": "text-embedding-3-large", "prompt": "Methods for tracing dataset provenance and legal implications in AI training data ", "output": [["Methods for tracing dataset provenance and legal implications in AI training data ", "Title: The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI\n\nAbstract: The race to train language models on vast, diverse, and inconsistently documented datasets has raised pressing concerns about the legal and ethical risks for practitioners. To remedy these practices threatening data transparency and understanding, we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use. Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data. This points to a deepening divide in the types of data that are made available under different license conditions, and heightened implications for jurisdictional legal interpretations of copyright and fair use. We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs. As a contribution to ongoing improvements in dataset transparency and responsible use, we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections: www.dataprovenance.org."]], "corpus": "arxiv"}
|
25 |
{"tstamp": 1722306358.068, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722306357.4875, "finish": 1722306358.068, "ip": "", "conv_id": "ff92d927e7a1440f91ca6ddae1fb40a5", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "Methods for tracing dataset provenance and legal implications in AI training data ", "output": [["Methods for tracing dataset provenance and legal implications in AI training data ", "Title: The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI\n\nAbstract: The race to train language models on vast, diverse, and inconsistently documented datasets has raised pressing concerns about the legal and ethical risks for practitioners. To remedy these practices threatening data transparency and understanding, we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use. Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data. This points to a deepening divide in the types of data that are made available under different license conditions, and heightened implications for jurisdictional legal interpretations of copyright and fair use. We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs. As a contribution to ongoing improvements in dataset transparency and responsible use, we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections: www.dataprovenance.org."]], "corpus": "arxiv"}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
23 |
{"tstamp": 1722306330.1591, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722306329.8317, "finish": 1722306330.1591, "ip": "", "conv_id": "940948f56e9c4f5297cb17f69935bd8b", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "when did scotland last qualify for world cup", "output": [["when did scotland last qualify for world cup", "Scotland at the FIFA World Cup\n\nThe World Cup consists of two parts, the qualification phase and the final phase (officially called the World Cup Finals). The qualification phase, which currently take place over the three years preceding the Finals, is used to determine which teams qualify for the Finals. The current format of the Finals involves 32 teams competing at venues within the host nation (or nations) over a period of about a month. The World Cup Finals is the most widely viewed sporting event in the world, with an estimated 715.1 million people watching the 2006 tournament final.\nHistory\n1930s\nScotland did not compete in the first three World Cup competitions, in 1930, 1934 and 1938. Because of a dispute with FIFA over \"broken-time\" payments to players, the SFA, with the football associations of England, Ireland and Wales, withdrew from FIFA in 1928, and did not rejoin as a permanent member until 1946.\n1950s\nThe readmission of the Scottish Football Association to FIFA in 1946 meant that Scotland were now eligible to enter the 1950 FIFA World Cup. FIFA advised that places would be awarded to the top two teams in the 1949–50 British Home Championship, but the SFA announced that Scotland would attend the finals only if Scotland won the competition. Scotland won their first two matches, but a 1–0 home defeat by England meant that the Scots finished as runners-up. The SFA stood by this proclamation, despite pleas to the contrary by the Scotland players, supported by England captain Billy Wright and the other England players. Sportswriter Brian Glanville later described the SFA as showing \"baffling insularity and pique\" in their refusal to send a team to the World Cup."]], "corpus": "wikipedia"}
|
24 |
{"tstamp": 1722306358.068, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722306357.4875, "finish": 1722306358.068, "ip": "", "conv_id": "be3fc968a3aa487ba8db65457d91fec3", "model_name": "text-embedding-3-large", "prompt": "Methods for tracing dataset provenance and legal implications in AI training data ", "output": [["Methods for tracing dataset provenance and legal implications in AI training data ", "Title: The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI\n\nAbstract: The race to train language models on vast, diverse, and inconsistently documented datasets has raised pressing concerns about the legal and ethical risks for practitioners. To remedy these practices threatening data transparency and understanding, we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use. Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data. This points to a deepening divide in the types of data that are made available under different license conditions, and heightened implications for jurisdictional legal interpretations of copyright and fair use. We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs. As a contribution to ongoing improvements in dataset transparency and responsible use, we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections: www.dataprovenance.org."]], "corpus": "arxiv"}
|
25 |
{"tstamp": 1722306358.068, "task_type": "retrieval", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722306357.4875, "finish": 1722306358.068, "ip": "", "conv_id": "ff92d927e7a1440f91ca6ddae1fb40a5", "model_name": "BAAI/bge-large-en-v1.5", "prompt": "Methods for tracing dataset provenance and legal implications in AI training data ", "output": [["Methods for tracing dataset provenance and legal implications in AI training data ", "Title: The Data Provenance Initiative: A Large Scale Audit of Dataset Licensing & Attribution in AI\n\nAbstract: The race to train language models on vast, diverse, and inconsistently documented datasets has raised pressing concerns about the legal and ethical risks for practitioners. To remedy these practices threatening data transparency and understanding, we convene a multi-disciplinary effort between legal and machine learning experts to systematically audit and trace 1800+ text datasets. We develop tools and standards to trace the lineage of these datasets, from their source, creators, series of license conditions, properties, and subsequent use. Our landscape analysis highlights the sharp divides in composition and focus of commercially open vs closed datasets, with closed datasets monopolizing important categories: lower resource languages, more creative tasks, richer topic variety, newer and more synthetic training data. This points to a deepening divide in the types of data that are made available under different license conditions, and heightened implications for jurisdictional legal interpretations of copyright and fair use. We also observe frequent miscategorization of licenses on widely used dataset hosting sites, with license omission of 70%+ and error rates of 50%+. This points to a crisis in misattribution and informed use of the most popular datasets driving many recent breakthroughs. As a contribution to ongoing improvements in dataset transparency and responsible use, we release our entire audit, with an interactive UI, the Data Provenance Explorer, which allows practitioners to trace and filter on data provenance for the most popular open source finetuning data collections: www.dataprovenance.org."]], "corpus": "arxiv"}
|
26 |
+
{"tstamp": 1722306925.7354, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722306925.3084, "finish": 1722306925.7354, "ip": "", "conv_id": "f027bc84f35a4d748c67b44a2435bbe4", "model_name": "voyage-multilingual-2", "prompt": "What is a transformer?", "output": [["What is a transformer?", "Title: Formal Algorithms for Transformers\n\nAbstract: This document aims to be a self-contained, mathematically precise overview of transformer architectures and algorithms (*not* results). It covers what transformers are, how they are trained, what they are used for, their key architectural components, and a preview of the most prominent models. The reader is assumed to be familiar with basic ML terminology and simpler neural network architectures such as MLPs."]], "corpus": "arxiv"}
|
27 |
+
{"tstamp": 1722306971.5819, "task_type": "retrieval", "type": "chat", "model": "mixedbread-ai/mxbai-embed-large-v1", "gen_params": {}, "start": 1722306971.4007, "finish": 1722306971.5819, "ip": "", "conv_id": "a55ff04a15fd4107ac10365fbb0eb9a2", "model_name": "mixedbread-ai/mxbai-embed-large-v1", "prompt": "where did the british empire control an entire continent", "output": [["where did the british empire control an entire continent", "British Empire\n\nThe British Empire comprised the dominions, colonies, protectorates, mandates, and other territories ruled or administered by the United Kingdom and its predecessor states. It began with the overseas possessions and trading posts established by England in the late 16th and early 17th centuries. At its height in the 19th and early 20th centuries, it was the largest empire in history and, for a century, was the foremost global power. By 1913, the British Empire held sway over 412 million people, of the world population at the time, and by 1920, it covered , of the Earth's total land area. As a result, its constitutional, legal, linguistic, and cultural legacy is widespread. At the peak of its power, it was described as \"the empire on which the sun never sets\", as the sun was always shining on at least one of its territories.\nDuring the Age of Discovery in the 15th and 16th centuries, Portugal and Spain pioneered European exploration of the globe, and in the process established large overseas empires. Envious of the great wealth these empires generated, England, France, and the Netherlands began to establish colonies and trade networks of their own in the Americas and Asia. A series of wars in the 17th and 18th centuries with the Netherlands and France left England (Britain, following the 1707 Act of Union with Scotland) the dominant colonial power in North America. Britain became a major power in the Indian subcontinent after the East India Company's conquest of Mughal Bengal at the Battle of Plassey in 1757."]], "corpus": "wikipedia"}
|
28 |
+
{"tstamp": 1722306971.5819, "task_type": "retrieval", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722306971.4007, "finish": 1722306971.5819, "ip": "", "conv_id": "a6afd35e6fa54368be503cb1c3cf43ea", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": "where did the british empire control an entire continent", "output": [["where did the british empire control an entire continent", "Territorial evolution of the British Empire\n\nThe territorial evolution of the British Empire is considered to have begun with the foundation of the English colonial empire in the late 16th century. Since then, many territories around the world have been under the control of the United Kingdom or its predecessor states.\nWhen the Kingdom of Great Britain was formed in 1707 by the union of the Kingdoms of Scotland and England, the latter country's colonial possessions passed to the new state. Similarly, when Great Britain was united with the Kingdom of Ireland in 1801 to form the United Kingdom, control over its colonial possessions passed to the latter state. Collectively, these territories are referred to as the British Empire. When much of Ireland gained independence in 1922 as the Irish Free State, the other territories of the empire remained under the control of the United Kingdom.\nFrom 1714 to 1837, the British throne was held by a series of kings who were also the rulers of the German state of Hanover. However, this was purely a personal union, with Hanover maintaining its political independence otherwise, and so it is not usually considered to have formed part of the British Empire."]], "corpus": "wikipedia"}
|
29 |
+
{"tstamp": 1722306986.0121, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722306985.8309, "finish": 1722306986.0121, "ip": "", "conv_id": "734d33bae026481cab60cb1e123a36b4", "model_name": "GritLM/GritLM-7B", "prompt": "Is attention all I need?", "output": [["Is attention all I need?", "Title: Attention Is All You Need\n\nAbstract: The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.8 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data."]], "corpus": "arxiv"}
|
30 |
+
{"tstamp": 1722307003.9939, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722307003.6084, "finish": 1722307003.9939, "ip": "", "conv_id": "3c77700a297d4a26b9466b34301c929b", "model_name": "voyage-multilingual-2", "prompt": "Is attention all I need?", "output": [["Is attention all I need?", "Title: Visual Attention and its Intimate Links to Spatial Cognition\n\nAbstract: It is almost universal to regard attention as the facility that permits an agent, human or machine, to give priority processing resources to relevant stimuli while ignoring the irrelevant. The reality of how this might manifest itself throughout all the forms of perceptual and cognitive processes possessed by humans, however, is not as clear. Here we examine this reality with a broad perspective in order to highlight the myriad ways that attentional processes impact both perception and cognition. The paper concludes by showing two real world problems that exhibit sufficient complexity to illustrate the ways in which attention and cognition connect. These then point to new avenues of research that might illuminate the overall cognitive architecture of spatial cognition."]], "corpus": "arxiv"}
|
31 |
+
{"tstamp": 1722307006.8942, "task_type": "retrieval", "type": "chat", "model": "voyage-multilingual-2", "gen_params": {}, "start": 1722307006.267, "finish": 1722307006.8942, "ip": "", "conv_id": "fdbaddd493b84cc89357c4480e2d727d", "model_name": "voyage-multilingual-2", "prompt": "The effect of Long pimple ", "output": [["The effect of Long pimple ", "Pimple\n\nA pimple or zit is a kind of comedo that results from excess sebum and dead skin cells getting trapped in the pores of the skin. In its aggravated state, it may evolve into a pustule or papule. Pimples can be treated by acne medications, antibiotics, and anti-inflammatories prescribed by a physician, or various over the counter remedies purchased at a pharmacy.\nCauses\nSebaceous glands inside the pore of the skin produce sebum. When the outer layers of skin shed (a natural and continuous process, normally), dead skin and oily sebum left behind may bond together and form a blockage of the sebaceous gland at the base of the skin. This is most common when the skin becomes thicker at puberty. The sebaceous gland continues to produce sebum, which builds up behind the blockage, allowing bacteria to grow in the area, including the species Staphylococcus aureus and Cutibacterium acnes, which causes inflammation and infection. Other causes of pimples include family history, stress, fluctuations in hormone levels, hair and skincare products, medication side effects, and un-diagnosed or underlying medical conditions. Pimples can be part of the presentation of rosacea."]], "corpus": "wikipedia"}
|
32 |
+
{"tstamp": 1722307006.8942, "task_type": "retrieval", "type": "chat", "model": "text-embedding-3-large", "gen_params": {}, "start": 1722307006.267, "finish": 1722307006.8942, "ip": "", "conv_id": "86047e9633f3430da6d888564b950f65", "model_name": "text-embedding-3-large", "prompt": "The effect of Long pimple ", "output": [["The effect of Long pimple ", "Table tennis grips and playing styles\n\nShort pips hitter This penhold style utilizes a short pips out rubber, usually on only one side of the racket with no backside rubber. Short pips hitters play over the table, hitting the ball very flatly as soon as it bounces off the table with the pips being largely unaffected by (or even negating) the majority of the opponent's spin. The advantage of hitting the ball flatly is that all of the player's power is transferred into the linear velocity of the ball, without expending any energy on creating spin. This means that the shots are very fast, and typically hard to block for the inexperienced player. However, the lack of a topspin Magnus effect means that the shots are harder to place on the table since they don't follow a downward arc trajectory, and the lack of angular momentum on the ball means that shots are less consistent and more sensitive to small amounts of wind or air pressure, often giving a knuckle ball effect.\nNotable penhold short pips hitters include 1996 Olympic Champion Liu Guoliang, 1985 and 1987 World Champion Jiang Jialiang, three-time Olympian Toshio Tasaki, Yang Ying, Kwak Bang-bang, Seok Eun-mi, He Zhi Wen, Lee Eun-hee, Wang Zeng Yi and Rory Cargill.\nShakehand playing styles"]], "corpus": "wikipedia"}
|