Scheduled Commit

data/clustering_battle-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -0,0 +1 @@
+{"tstamp": 1722523947.4992, "task_type": "clustering", "type": "rightvote", "models": ["", ""], "ip": "", "0_conv_id": "87d24144100b4178817631dcf0a099ad", "0_model_name": "BAAI/bge-large-en-v1.5", "0_prompt": ["Shanghai", "Beijing", "Shenzhen", "Hangzhou", "Seattle", "Boston", "New York", "San Francisco"], "0_ncluster": 2, "0_output": "", "0_ndim": "3D (press for 2D)", "0_dim_method": "PCA", "0_clustering_method": "KMeans", "1_conv_id": "7a4073742c284c25966dccad707db8cb", "1_model_name": "sentence-transformers/all-MiniLM-L6-v2", "1_prompt": ["Shanghai", "Beijing", "Shenzhen", "Hangzhou", "Seattle", "Boston", "New York", "San Francisco"], "1_ncluster": 2, "1_output": "", "1_ndim": "3D (press for 2D)", "1_dim_method": "PCA", "1_clustering_method": "KMeans"}

data/clustering_individual-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -0,0 +1,2 @@
+{"tstamp": 1722523906.041, "task_type": "clustering", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722523893.5336, "finish": 1722523906.041, "ip": "", "conv_id": "87d24144100b4178817631dcf0a099ad", "model_name": "BAAI/bge-large-en-v1.5", "prompt": ["Shanghai", "Beijing", "Shenzhen", "Hangzhou", "Seattle", "Boston", "New York", "San Francisco"], "ncluster": 2, "output": "", "ndim": "3D (press for 2D)", "dim_method": "PCA", "clustering_method": "KMeans"}
+{"tstamp": 1722523906.041, "task_type": "clustering", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722523893.5336, "finish": 1722523906.041, "ip": "", "conv_id": "7a4073742c284c25966dccad707db8cb", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": ["Shanghai", "Beijing", "Shenzhen", "Hangzhou", "Seattle", "Boston", "New York", "San Francisco"], "ncluster": 2, "output": "", "ndim": "3D (press for 2D)", "dim_method": "PCA", "clustering_method": "KMeans"}

data/retrieval_battle-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -5,3 +5,5 @@
 {"tstamp": 1722523333.6616, "task_type": "retrieval", "type": "rightvote", "models": ["", ""], "ip": "", "0_conv_id": "6a0e031567b94ddd9724312eb435daa6", "0_model_name": "BM25", "0_prompt": "what is the difference between what time and when?", "0_output": [["what is the difference between what time and when?", "Stackoverflow Stackexchange\n\nQ: what is the difference between O(nk) and O(n+k) in time complexity? In big O notation of time complexity in algorithmic analysis, when an algorithm depends on n and k, what is the difference between these two notations. \nAlso pls help in the notation to use if there is a nested loop with outer loop running n times and inner loop running k times ?\n\nA: O(nk):\nfor( i=0; i<n; i++ ) {\n   for( j=0; j<k; j++ )\n   {}\n}\n\nO(n+k):\nfor( i=0; i<n; i++ )\n{}\n\nfor( j=0; j<k; j++ )\n{}\n\n"]], "0_corpus": "stackexchange", "1_conv_id": "43787c9534da42b1ac358069ebd04c1c", "1_model_name": "Salesforce/SFR-Embedding-2_R", "1_prompt": "what is the difference between what time and when?", "1_output": [["what is the difference between what time and when?", "Ell Stackexchange\n\nQ: \"What time...?\" or \"At what time...?\" - what is more grammatically correct? This question may sound silly, but it has been bugging me for years. \nIf I ask a question about a precise point in time, should I say \"What time...\" or \"At what time...\"?\nFor example,\n\nAt what time does Billy arrive at the swimming pool?\n\nor\n\nWhat time does Billy arrive at the swimming pool?\n\n\nA: As user070221 notes, both sentences are commonly used in American English.  In some formal speech and writing, \"At what time\" is more acceptable than \"When\" or \"What time\", especially when \"a precise point in time\" is being requested. \nI am an American who grew up in a town with many native speakers of Spanish.  To my ear, both examples in the original post sound like overly literal translations of \"¿A qué hora?\"  To my ear, \"When\" is a more natural way of saying this in English:\n\nWhen does Billy arrive at the swimming pool?  \n\n"]], "1_corpus": "stackexchange"}
 {"tstamp": 1722523807.796, "task_type": "retrieval", "type": "rightvote", "models": ["", ""], "ip": "", "0_conv_id": "dfe9f28cf60c4069ab597a643b9f983c", "0_model_name": "jinaai/jina-embeddings-v2-base-en", "0_prompt": "Which test was devised to determine whether robots can think?", "0_output": [["Which test was devised to determine whether robots can think?", "Cognitive test\n\nCognitive tests are assessments of the cognitive capabilities of humans and other animals. Tests administered to humans include various forms of IQ tests; those administered to animals include the mirror test (a test of visual self-awareness) and the T maze test (which tests learning ability). Such testing is used in psychology and psychometrics, as well as other fields studying human and animal intelligence."]], "0_corpus": "wikipedia", "1_conv_id": "aa699ca67c654ea1861add38516ac50b", "1_model_name": "nomic-ai/nomic-embed-text-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": 1722523837.949, "task_type": "retrieval", "type": "tievote", "models": ["", ""], "ip": "", "0_conv_id": "c02265c3c8ac4f068a034d5d4971c7f2", "0_model_name": "voyage-multilingual-2", "0_prompt": "what model are you?", "0_output": [["what model are you?", "Model (disambiguation)\n\nA model is an informative representation of an object, person or system.\nModel may also refer to:\nFilm and television\nModel (TV series), a 1997 South Korean television series\nThe Model (film), a 2016 Danish thriller drama film\nModels, a 1999 Austrian drama film by Ulrich Seidl\nLiterature\nModel (manhwa), a 1999 series by Lee So-young\nThe Model, a 2005 novel by Lars Saabye Christensen\nMusic\nModel (band), a Turkish rock band\nModels (band), an Australian rock band\nThe Models, an English punk rock band\n\"Model\" (Gulddreng song), 2016\n\"Das Model\", a 1978 song by Kraftwerk\nModel (album), a 2024 album by Wallows\n\"Model\", a 1994 song by Avail from Dixie\n\"Model\", a 1991 song by Simply Red from Stars\nPeople\nModel (surname), a surname frequently of Central European and occasionally English origins\nThe Model (wrestler), ring name of Rick Martel (born 1956)\nEddie Taubensee (born 1968), baseball player nicknamed \"The Model\"\nPlaces\nModel, Colorado, an unincorporated town in the United States\nModel, Masovian Voivodeship, a village in east-central Poland\nOther uses\nModel (mathematical logic), a structure that satisfies a given system of axioms\nMachine learning model, a special type of a mathematical model in the context of machine learning\nMODEL or Movement for Democracy in Liberia, a rebel group\nModel (person), a role or occupation"]], "0_corpus": "wikipedia", "1_conv_id": "bbe903b763244601871fd1e4ea0525b5", "1_model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "1_prompt": "what model are you?", "1_output": [["what model are you?", "Choice\n\nA choice is the range of different things from which a being can choose.  The arrival at a choice may incorporate motivators and models."]], "1_corpus": "wikipedia"}
+{"tstamp": 1722523841.5582, "task_type": "retrieval", "type": "rightvote", "models": ["", ""], "ip": "", "0_conv_id": "5fbfc4c28f704d3e8f79700e67543616", "0_model_name": "GritLM/GritLM-7B", "0_prompt": "how to fuck a woman", "0_output": [["how to fuck a woman", "The Perfumed Garden\n\nUgly women are repulsive but also those who are overly raucous and have a loud laugh. A woman who is often seen joking and fooling around may be promiscuous. Also listed as disagreeable traits are revealing their husband's secrets, delighting in other's misfortune, pointing out other's shortcomings, busy-bodies, shrews, talkers, gossips, the lazy, harridans, the hysteric, the nag and the pilfering slag.\nChapter 5: Sexual Intercourse\nIt is recommended that a man should not eat or drink too much before having sex and that foreplay is necessary in order to excite the woman. When finished the man should not rush to leave and should do so on his right hand side.\nChapter 6: Sexual Technique\nThis chapter provides instructions on foreplay, specifying that it should include cunnilingus. The importance of the woman's enjoyment and climax are stressed, as are a number of steps to be taken to avoid injury or infection. Concerning sexual positions it is said that all are permissible (but Khawam's translation adds the words \"except in her rear end\" i.e. anal sex). Eleven positions are then listed, six with the woman on her back, one from behind, two with one or both on their sides, one over furniture and one hanging from a tree.\nChapter 7: The Harmful Effects of Intercourse"]], "0_corpus": "wikipedia", "1_conv_id": "64cdf389bebc41d991e2ad31fcafa671", "1_model_name": "sentence-transformers/all-MiniLM-L6-v2", "1_prompt": "how to fuck a woman", "1_output": [["how to fuck a woman", "How to Please a Woman\n\nHow to Please a Woman is a 2022 Australian comedy-drama film directed by Renée Webster, starring Sally Phillips, Caroline Brazier, Erik Thomson, Tasma Walton and Alexander England."]], "1_corpus": "wikipedia"}
+{"tstamp": 1722523865.2136, "task_type": "retrieval", "type": "leftvote", "models": ["", ""], "ip": "", "0_conv_id": "0076b2f0f7b541fb88a7132b3c764c4f", "0_model_name": "intfloat/e5-mistral-7b-instruct", "0_prompt": "how to kill people", "0_output": [["how to kill people", "List of types of killing\n\nSiblicide – the killing of an infant individual by their close relatives (full or half siblings).\nSororicide – the act of killing one's sister ( \"sister\").\nUxoricide – the act of killing one's wife ( \"wife\").\nKilling of others\nAmicicide – the act of killing a friend ( \"friend\").\nAndrocide – the systematic killing of men.\nAssassination – the act of killing a prominent person for either political, religious, or monetary reasons.\nCapital punishment – the judicial killing of a human being for crimes.\nCasualty – death (or injury) in wartime.\nCollateral damage – Incidental killing of persons during a military attack that were not the object of attack.\nDemocide or populicide – the murder of any person or people by a government.\nExtrajudicial killing – killing by government forces without due process. See also Targeted killing.\nEuthanasia or mercy killing – the killing of any being with compassionate reasoning; e.g., significant injury or disease.\nFamiliaricide in commutatione eius possessio – the act of killing a family for their property and/or possessions (from  \"of a household\";  \"in exchange for\"; and  \"a possession or property\").\nFemicide, gynecide, gynaecide, or gynocide – the systematic killing of women.\nFeticide – the killing of an embryo or fetus.\nFragging - the act of killing a fellow soldier."]], "0_corpus": "wikipedia", "1_conv_id": "ada677e671634ec99987bff3781499b0", "1_model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "1_prompt": "how to kill people", "1_output": [["how to kill people", "The Death of Adolf Hitler\n\nAuthor Soviet journalist Lev Bezymenski (1920–2007), the son of poet Aleksandr Bezymensky, served as an interpreter in the Battle of Berlin under Marshal Zhukov. Early on 1 May 1945, he translated a letter from Goebbels and Bormann announcing Hitler's death. Bezymenski authored several works about the Nazi era.\nContent\nThe book begins with an overview of the Battle of Berlin and its aftermath, including a reproduction of the purported Soviet autopsy report of Hitler's body. Bezymenski states that the bodies of Hitler and Braun were \"the most seriously disfigured of all thirteen corpses\" examined. The appendix summarizes the discovery of the Goebbels family's corpses and includes further forensic reports. On why the autopsy reports were not released earlier, Bezymenski says:Not because of doubts as to the credibility of the experts. ... Those who were involved in the investigation remember that other considerations played a far larger role. First, it was resolved not to publish the results of the forensic-medical report but to \"hold it in reserve\" in case someone might try to slip into the role of \"the Führer saved by a miracle.\" Secondly, it was resolved to continue the investigations in order to exclude any possibility of error or deliberate deception.\nThe Death of Adolf Hitler"]], "1_corpus": "wikipedia"}

data/retrieval_individual-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -16,3 +16,7 @@
 {"tstamp": 1722523816.3844, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1722523815.9563, "finish": 1722523816.3844, "ip": "", "conv_id": "bbe903b763244601871fd1e4ea0525b5", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "what model are you?", "output": [["what model are you?", "Choice\n\nA choice is the range of different things from which a being can choose.  The arrival at a choice may incorporate motivators and models."]], "corpus": "wikipedia"}
 {"tstamp": 1722523831.7471, "task_type": "retrieval", "type": "chat", "model": "GritLM/GritLM-7B", "gen_params": {}, "start": 1722523831.3895, "finish": 1722523831.7471, "ip": "", "conv_id": "5fbfc4c28f704d3e8f79700e67543616", "model_name": "GritLM/GritLM-7B", "prompt": "how to fuck a woman", "output": [["how to fuck a woman", "The Perfumed Garden\n\nUgly women are repulsive but also those who are overly raucous and have a loud laugh. A woman who is often seen joking and fooling around may be promiscuous. Also listed as disagreeable traits are revealing their husband's secrets, delighting in other's misfortune, pointing out other's shortcomings, busy-bodies, shrews, talkers, gossips, the lazy, harridans, the hysteric, the nag and the pilfering slag.\nChapter 5: Sexual Intercourse\nIt is recommended that a man should not eat or drink too much before having sex and that foreplay is necessary in order to excite the woman. When finished the man should not rush to leave and should do so on his right hand side.\nChapter 6: Sexual Technique\nThis chapter provides instructions on foreplay, specifying that it should include cunnilingus. The importance of the woman's enjoyment and climax are stressed, as are a number of steps to be taken to avoid injury or infection. Concerning sexual positions it is said that all are permissible (but Khawam's translation adds the words \"except in her rear end\" i.e. anal sex). Eleven positions are then listed, six with the woman on her back, one from behind, two with one or both on their sides, one over furniture and one hanging from a tree.\nChapter 7: The Harmful Effects of Intercourse"]], "corpus": "wikipedia"}
 {"tstamp": 1722523831.7471, "task_type": "retrieval", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722523831.3895, "finish": 1722523831.7471, "ip": "", "conv_id": "64cdf389bebc41d991e2ad31fcafa671", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": "how to fuck a woman", "output": [["how to fuck a woman", "How to Please a Woman\n\nHow to Please a Woman is a 2022 Australian comedy-drama film directed by Renée Webster, starring Sally Phillips, Caroline Brazier, Erik Thomson, Tasma Walton and Alexander England."]], "corpus": "wikipedia"}
+{"tstamp": 1722523855.6401, "task_type": "retrieval", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722523855.3353, "finish": 1722523855.6401, "ip": "", "conv_id": "6ca957400f2a4f0c9f9f0a98b9c88c99", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "prompt": "who was the best person ever?", "output": [["who was the best person ever?", "The Greatest of All Time\n\nThe Greatest of All Time (also marketed as GOAT) is an upcoming Indian Tamil-language science fiction action film directed by Venkat Prabhu and produced by AGS Entertainment. The film stars Vijay in dual roles, leading an ensemble cast which includes Prashanth, Prabhu Deva, Mohan, Jayaram, Sneha, Laila, Ajmal Ameer, Meenakshi Chaudhary, Vaibhav, Yogi Babu, Premgi Amaren and Yugendran.\nThe film was officially announced in May 2023 under the tentative title Thalapathy 68, as it is Vijay's 68th film in a leading role, and the official title was announced that December. Principal photography commenced in October 2023. It was shot in several locations including Chennai, Thailand, Hyderabad, Sri Lanka, Pondicherry, Thiruvananthapuram, Russia and United States. The film has music composed by Yuvan Shankar Raja, cinematography handled by Siddhartha Nuni and editing by Venkat Raajen.\nThe Greatest of All Time is scheduled to release worldwide on 5 September 2024 in theatres.\nCast\nAdditionally, late actor Vijayakanth and Y. G. Mahendran will make cameo appearances, with the makers using artificial intelligence for the former.\nProduction"]], "corpus": "wikipedia"}
+{"tstamp": 1722523855.6401, "task_type": "retrieval", "type": "chat", "model": "Salesforce/SFR-Embedding-2_R", "gen_params": {}, "start": 1722523855.3353, "finish": 1722523855.6401, "ip": "", "conv_id": "c1389c5c097b4b728b727e898c568942", "model_name": "Salesforce/SFR-Embedding-2_R", "prompt": "who was the best person ever?", "output": [["who was the best person ever?", "Time 100: The Most Important People of the Century\n\nTime 100: The Most Important People of the Century''' is a compilation of the 20th century's 100 most influential people, published in Time magazine across five issues in 1998 and 1999.\nThe idea for such a list started on February 1, 1998, with a debate at a symposium in Hanoi, Vietnam. The panel participants were former CBS Evening News anchor Dan Rather, historian Doris Kearns Goodwin, former New York governor Mario Cuomo, then–Stanford Provost Condoleezza Rice, publisher Irving Kristol, and Time managing editor Walter Isaacson.\nIn a separate issue on December 31, 1999, Time recognized Albert Einstein as the Person of the Century.\nList categoriesTime cited 20 selections in each of five categories: Leaders & Revolutionaries, Artists & Entertainers, Builders & Titans, Scientists & Thinkers, and Heroes & Icons.\nPerson of the Century\nOf the 100 chosen, Albert Einstein was chosen as the Person of the Century, on the grounds that he was the preeminent scientist in a century dominated by science. The editors of Time believed the 20th century \"will be remembered foremost for its science and technology\", and Einstein \"serves as a symbol of all the scientists—such as Fermi, Heisenberg, Bohr, Richard Feynman, ...who built upon his work\"."]], "corpus": "wikipedia"}
+{"tstamp": 1722523857.8576, "task_type": "retrieval", "type": "chat", "model": "intfloat/e5-mistral-7b-instruct", "gen_params": {}, "start": 1722523856.9882, "finish": 1722523857.8576, "ip": "", "conv_id": "0076b2f0f7b541fb88a7132b3c764c4f", "model_name": "intfloat/e5-mistral-7b-instruct", "prompt": "how to kill people", "output": [["how to kill people", "List of types of killing\n\nSiblicide – the killing of an infant individual by their close relatives (full or half siblings).\nSororicide – the act of killing one's sister ( \"sister\").\nUxoricide – the act of killing one's wife ( \"wife\").\nKilling of others\nAmicicide – the act of killing a friend ( \"friend\").\nAndrocide – the systematic killing of men.\nAssassination – the act of killing a prominent person for either political, religious, or monetary reasons.\nCapital punishment – the judicial killing of a human being for crimes.\nCasualty – death (or injury) in wartime.\nCollateral damage – Incidental killing of persons during a military attack that were not the object of attack.\nDemocide or populicide – the murder of any person or people by a government.\nExtrajudicial killing – killing by government forces without due process. See also Targeted killing.\nEuthanasia or mercy killing – the killing of any being with compassionate reasoning; e.g., significant injury or disease.\nFamiliaricide in commutatione eius possessio – the act of killing a family for their property and/or possessions (from  \"of a household\";  \"in exchange for\"; and  \"a possession or property\").\nFemicide, gynecide, gynaecide, or gynocide – the systematic killing of women.\nFeticide – the killing of an embryo or fetus.\nFragging - the act of killing a fellow soldier."]], "corpus": "wikipedia"}
+{"tstamp": 1722523857.8576, "task_type": "retrieval", "type": "chat", "model": "Alibaba-NLP/gte-Qwen2-7B-instruct", "gen_params": {}, "start": 1722523856.9882, "finish": 1722523857.8576, "ip": "", "conv_id": "ada677e671634ec99987bff3781499b0", "model_name": "Alibaba-NLP/gte-Qwen2-7B-instruct", "prompt": "how to kill people", "output": [["how to kill people", "The Death of Adolf Hitler\n\nAuthor Soviet journalist Lev Bezymenski (1920–2007), the son of poet Aleksandr Bezymensky, served as an interpreter in the Battle of Berlin under Marshal Zhukov. Early on 1 May 1945, he translated a letter from Goebbels and Bormann announcing Hitler's death. Bezymenski authored several works about the Nazi era.\nContent\nThe book begins with an overview of the Battle of Berlin and its aftermath, including a reproduction of the purported Soviet autopsy report of Hitler's body. Bezymenski states that the bodies of Hitler and Braun were \"the most seriously disfigured of all thirteen corpses\" examined. The appendix summarizes the discovery of the Goebbels family's corpses and includes further forensic reports. On why the autopsy reports were not released earlier, Bezymenski says:Not because of doubts as to the credibility of the experts. ... Those who were involved in the investigation remember that other considerations played a far larger role. First, it was resolved not to publish the results of the forensic-medical report but to \"hold it in reserve\" in case someone might try to slip into the role of \"the Führer saved by a miracle.\" Secondly, it was resolved to continue the investigations in order to exclude any possibility of error or deliberate deception.\nThe Death of Adolf Hitler"]], "corpus": "wikipedia"}

data/sts_battle-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -0,0 +1 @@
+{"tstamp": 1722524028.4017, "task_type": "sts", "type": "bothbadvote", "models": ["", ""], "ip": "", "0_conv_id": "73f80a1329a74718bf37b09aafcddd9e", "0_model_name": "BAAI/bge-large-en-v1.5", "0_txt0": "hello", "0_txt1": "good morning", "0_txt2": "早上好", "0_output": "", "1_conv_id": "93eab62523e84cdc9724270434510fe7", "1_model_name": "sentence-transformers/all-MiniLM-L6-v2", "1_txt0": "hello", "1_txt1": "good morning", "1_txt2": "早上好", "1_output": ""}

data/sts_individual-d39d676b-bc57-4c8b-9e61-c980be2f2db2.jsonl

@@ -0,0 +1,2 @@
+{"tstamp": 1722523994.6594, "task_type": "sts", "type": "chat", "model": "BAAI/bge-large-en-v1.5", "gen_params": {}, "start": 1722523994.3318, "finish": 1722523994.6594, "ip": "", "conv_id": "73f80a1329a74718bf37b09aafcddd9e", "model_name": "BAAI/bge-large-en-v1.5", "txt0": "hello", "txt1": "good morning", "txt2": "早上好", "output": ""}
+{"tstamp": 1722523994.6594, "task_type": "sts", "type": "chat", "model": "sentence-transformers/all-MiniLM-L6-v2", "gen_params": {}, "start": 1722523994.3318, "finish": 1722523994.6594, "ip": "", "conv_id": "93eab62523e84cdc9724270434510fe7", "model_name": "sentence-transformers/all-MiniLM-L6-v2", "txt0": "hello", "txt1": "good morning", "txt2": "早上好", "output": ""}