Datasets:

G-reen
/

fastdetector-val-stat-val

original string	prompt dict	topic string	format string	response_0 string	response_1 string	generator_model string	generation_params string	final_response string	jaccard_1 float64	jaccard_2 float64	levenshtein float64	softngram float64	cosdist float64	bertscore_precision float64	bertscore_recall float64	bertscore float64	moverscore float64	reranker float64	original_perplexity_llama_instruct float64	original_entropy_llama_instruct float64	original_topp_outlier_llama_instruct float64	original_topk_outlier_llama_instruct float64	original_fastdetectgpt_llama_instruct float64	original_perplexity_llama_base float64	original_entropy_llama_base float64	original_topp_outlier_llama_base float64	original_topk_outlier_llama_base float64	original_fastdetectgpt_llama_base float64	final_response_perplexity_llama_instruct float64	final_response_entropy_llama_instruct float64	final_response_topp_outlier_llama_instruct float64	final_response_topk_outlier_llama_instruct float64	final_response_fastdetectgpt_llama_instruct float64	final_response_perplexity_llama_base float64	final_response_entropy_llama_base float64	final_response_topp_outlier_llama_base float64	final_response_topk_outlier_llama_base float64	final_response_fastdetectgpt_llama_base float64	original_binoculars float64	final_response_binoculars float64	original_editlens_score_roberta_large float64	final_response_editlens_bucket_roberta_large int64	final_response_editlens_score_roberta_large float64	original_editlens_bucket_llama_3_2_3b int64	original_editlens_score_llama_3_2_3b float64	final_response_editlens_bucket_llama_3_2_3b int64	final_response_editlens_score_llama_3_2_3b float64	original_editlens_score_giga_roberta float64	final_response_editlens_bucket_giga_roberta int64	final_response_editlens_score_giga_roberta float64	original_editlens_score_giga_llama_3_2_3b float64	final_response_editlens_bucket_giga_llama_3_2_3b int64	final_response_editlens_score_giga_llama_3_2_3b float64
The sharp changes in North Korea policy accomplished by Kim Dae Jung and the Clinton administration were immediately challenged by George W. Bush, within weeks of his inauguration in 2001. Seven years later, the Lee Myung Bak administration appears to think that a daunting rupture occurred between the ROK and the U.S.,...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nModify this text to be as long as possible. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose.\nOutput the full new text with no extra statements or commentations.\nBegin directly w...	Politics	Audio Transcript	The sharp changes in North Korea policy accomplished by Kim Dae Jung and the Clinton administration were immediately challenged by George W. Bush, within weeks of his inauguration in 2		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	The sharp changes in North Korea policy accomplished by Kim Dae Jung and the Clinton administration were immediately challenged by George W. Bush, within weeks of his inauguration in 2	0.976311	0.98823	16,312	0	0.281264	0.181242	0.35453	0.27814	0.86997	-6.875	15.954113	2.651362	0.051531	0.115724	-3.077315	11.599926	2.380524	0.040931	0.088928	-2.011218	29.267077	2.941019	0.088235	0.147059	-1.137848	23.791296	2.750884	0.058824	0.147059	-1.100297	0.819125	0.961063	0.020366	0	0.193622	0	0.012329	0	0.044977	0.002321	1	0.279174	0.002795	0	0.019557
We've designed most of the mission resources as workbooks intentionally. We felt these were the most convenient for people to use - they fit nicely in your hand, are still easy to write in, don't take up much room in a bag and even fit inside a Bible or Bible bag. Similarly, to keep the costs down, we've printed them i...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nRewrite the weakest section in this text, but do not make any modifications to any other section.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what ...	Software	Tutorial	To utilize the booklet creation feature of your software, follow these steps: 1. Open your document and select the Print option. 2. Choose your preferred printer and click on the Properties button. 3. Locate your printer’s booklet setting, often found under the Layout tab. You may need to click on the Booklet / Poste...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	To utilize the booklet creation feature of your software, follow these steps: 1. Open your document and select the Print option. 2. Choose your preferred printer and click on the Properties button. 3. Locate your printer’s booklet setting, often found under the Layout tab. You may need to click on the Booklet / Poster ...	0.767123	0.903315	1,268	0.17256	0.236995	0.117753	0.187594	0.154113	0.626954	-5.1875	13.001837	2.39735	0.043814	0.054124	-1.576063	10.133401	2.37227	0.023196	0.048969	0.555041	8.683636	2.066372	0.060606	0.075758	-0.554271	7.834282	2.093538	0.045455	0.060606	0.209153	0.867307	0.84926	0.018878	1	0.338787	0	0.013199	2	0.706336	0.006363	2	0.721789	0.003034	1	0.44303
Water scarcity is a recognized global problem, with demand for water projected to exceed supply by 40% by 2030. By the same year, half the world’s population will be living in areas of high water stress. Hotel companies have both a strong commercial and moral imperative for addressing water use. Cost is a clear factor:...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nWrite a 'few-shot' prompt using minimal, fragmented excerpts from this text as examples. The prompt should be designed to teach an LLM the pattern needed to generate the rest of the text from scratch.\nDo not output anything besides what you were requested to...	Industrial	Product Page	Prompt: "Water usage is a concern for many industries, including hospitals. Reasons include cost and morality. Cost: water can account for 10% of utility bills. Morality: water is scarce and should not be wasted, especially in areas where locals rely on the same supply. Companies like Aqua Tuta offer benefits such as ...	Certainly! Here's the generated text: * Reduced water bills: With efficient water systems, hotels can lower their water consumption, leading to significant savings on utility bills. * Positive brand image: Hotels that prioritize water conservation can enhance their reputation by showcasing their commitment to sustain...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	* Reduced water bills: With efficient water systems, hotels can lower their water consumption, leading to significant savings on utility bills. * Positive brand image: Hotels that prioritize water conservation can enhance their reputation by showcasing their commitment to sustainability and social responsibility. * Gov...	0.909804	0.994792	1,137	0.867267	0.456555	0.256292	0.273276	0.264882	0.775029	-1.5	17.824293	2.459334	0.074194	0.135484	-3.424144	14.113471	2.473045	0.067742	0.106452	-1.479563	4.914557	1.295333	0.04375	0.03125	-2.553169	5.362646	1.489941	0.03125	0.04375	-1.519101	0.977142	1.011185	0.017753	3	0.999592	0	0.010475	3	0.999971	0.035451	3	0.997436	0.003921	3	0.843615
You can delete a branch uploaded to Launchpad through the web UI. First find the branch in Launchpad. You can get a list of your branches at: The branch page will have a URL like: Then choose the red minus sign at the end of the branch name. If you are not able to delete the branch, you will be shown why not on the bra...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nTake inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human.\nOutput...	Software	FAQ	Frequently Asked Questions -------------------------- ### How do I delete a file on Google Drive? To delete a file on Google Drive, first locate the file in your Drive. You can see a list of all your files at [Google Drive website](http://drive.google.com). Once you've found the file, click on the three dots next to...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	### How do I delete a file on Google Drive? To delete a file on Google Drive, first locate the file in your Drive. You can see a list of all your files at [Google Drive website](http://drive.google.com). Once you've found the file, click on the three dots next to the file name and select "Remove" from the dropdown menu...	0.881081	0.969697	1,736	0.995185	0.639747	0.288236	0.237661	0.263816	0.797996	7.8125	14.549122	2.536721	0.038961	0.103896	-0.551233	11.77278	2.568486	0.012987	0.077922	0.427566	2.621635	1.052269	0.02765	0.004608	1.330956	2.616438	1.012863	0.020737	0.002304	0.824697	0.862414	0.741797	0.017015	1	0.265081	0	0.00646	1	0.50419	0.011043	3	0.977909	0.002476	3	0.967286
Philip Kerr is the New York Times bestselling author of the acclaimed Bernie Gunther novels, three of which—Field Gray, The Lady from Zagreb, and Prussian Blue—were finalists for the Edgar® Award for Best Novel. Kerr has also won several Shamus Awards and the British Crime Writers’ Association Ellis Peters Award for Hi...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nWrite a piece of text with a different topic that the author of this text might write about.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you h...	Literature	Audio Transcript	The Renaissance was a period of explosive growth and change in art, science, and culture, spanning the 14th to the 17th century. During this time, artists such as Leonardo da Vinci and Michelangelo pushed the boundaries of what was possible in painting and sculpture, while scientists like Galileo and Copernicus challe...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	The Renaissance was a period of explosive growth and change in art, science, and culture, spanning the 14th to the 17th century. During this time, artists such as Leonardo da Vinci and Michelangelo pushed the boundaries of what was possible in painting and sculpture, while scientists like Galileo and Copernicus challen...	0.929907	0.988351	3,912	1	0.84604	0.266305	0.285526	0.276043	0.787406	15.5625	11.682384	2.232533	0.056273	0.081181	-3.640885	8.611351	2.169233	0.031365	0.054428	0.271913	3.392827	1.058887	0.04607	0.01355	-2.41363	3.393107	1.30656	0.02981	0.01084	1.109495	0.825113	0.854844	0.1262	3	0.991836	0	0.143031	3	0.999984	0.003592	3	0.999925	0.002355	3	0.999993
A detailed insight into the angel numbers and the world they represent can reveal the knowledge of destiny – the path of life, with all the happiness and successes, or misfortunes and difficulties it carries along. Numbers trigger vibrations, and it is the Force that shapes our lives, and by following the same pattern,...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nWrite another piece of text that resembles the style and language of this text.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you have written.\...	Religion	Knowledge Article	Knowledge Article: A profound understanding of color energy and the universe it embodies can shed light upon the journey of life, encompassing moments of triumph and failure, ease and adversity. Colors radiate vibrations, and it is the Cosmic Force that guides our existence, shaping our destinies through consistent p...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	Knowledge Article: A profound understanding of color energy and the universe it embodies can shed light upon the journey of life, encompassing moments of triumph and failure, ease and adversity. Colors radiate vibrations, and it is the Cosmic Force that guides our existence, shaping our destinies through consistent pat...	0.931759	0.985816	4,836	0.99818	0.717168	0.224255	0.222628	0.223443	0.712076	10.8125	20.151442	2.895528	0.061503	0.116932	-1.80492	17.047748	2.796116	0.050114	0.107062	-0.660861	9.155706	2.400698	0.022838	0.068515	2.420335	8.878797	2.332239	0.016313	0.065253	1.924879	0.919743	0.79571	0.039913	3	0.999383	0	0.038	3	0.944266	0.005393	3	0.989066	0.002804	2	0.671019
Let me know, maybe it was a lucky shot. I check in 3 hours again. Unplug your shifter.... worked here I unplugged the shifter. That worked for 2 times in a row. check later this evening Did that too, doesn't work either.. When it worked, pls let me know I did reinstall the game, not working got the 0842, digital copy f...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nEdit this text by replacing exactly one to three sentences, without changing anything else.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you ha...	Games	Comment Section	Let me know if it was a lucky shot. I will check in 3 hours again. Unplugging the shifter worked for me two times in a row. Check later this evening. I also tried reinstalling the game, but it didn't solve the issue. Got the 0842 error code with my digital copy from the PS Store. After doing a fresh install, switching...	In the last few days, several players reported issues with their racing wheels on various games and platforms. Some found success by unplugging and replugging the shifter, updating the firmware, or changing USB ports. Others suspect defective units or compatibility problems. Here are some of their stories: * "Unplugg...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	In the last few days, several players reported issues with their racing wheels on various games and platforms. Some found success by unplugging and replugging the shifter, updating the firmware, or changing USB ports. Others suspect defective units or compatibility problems. Here are some of their stories: * "Unpluggin...	0.79257	0.902857	1,034	0.590406	0.154765	0.174514	0.181937	0.178242	0.647648	2.9375	72.015324	4.123702	0.061466	0.248227	-1.126224	47.338616	3.675261	0.06383	0.208038	-1.473245	18.637849	3.042473	0.039216	0.106443	0.902942	15.911869	2.839721	0.047619	0.092437	0.598753	0.821649	0.806738	0.009468	0	0.115859	0	0.016718	1	0.382916	0.015745	1	0.361814	0.008715	1	0.307781
In a beautiful small village of Bastar, Jhitku and Mitki met at a local fair and It was love at first sight ! Soon after, the word spread across the family and Jhitku's family had no objection, but Mitki's seven brothers were not ready for it. Mitki's seven brothers didn't wanted to leave alone and demanded Jitku to st...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nEdit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow.\nOutput the full new text with no extra statements or commentations.\nBegin di...	Literature	Creative Writing	In the picturesque hamlet of Bastar, Jhitku and Mitki crossed paths at a local festival, and it was love at first sight. As the news of their budding romance reached their families, Jhitku's relatives welcomed the match, whereas Mitki's seven brothers raised objections. They insisted that Jhitku should reside with the...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	In the picturesque hamlet of Bastar, Jhitku and Mitki crossed paths at a local festival, and it was love at first sight. As the news of their budding romance reached their families, Jhitku's relatives welcomed the match, whereas Mitki's seven brothers raised objections. They insisted that Jhitku should reside with them...	0.845304	0.960526	1,417	0.744762	0.25715	0.163055	0.13779	0.15061	0.592162	0.125	17.40432	2.777897	0.054705	0.098468	-0.730703	15.763598	2.679438	0.04814	0.087527	-0.752667	13.421491	2.27063	0.073563	0.110345	-3.392392	11.763297	2.246169	0.048276	0.094253	-2.31673	0.912344	0.959428	0.012112	1	0.34355	0	0.017545	2	0.606165	0.004737	1	0.385034	0.011611	1	0.462
As in most cartoons from the early Silly Symphonies, sight gags abound. I wonder if anyone laughed at the end of this cartoon. I gasped. It’s not to say that this is a terrible cartoon, but it’s just not up to par with what the cartoons preceding it have done. The basics of the short are that a lovely barmaid, Carmen, ...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nTake inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human.\nOutput...	Entertainment	User Review	User Review: Entertainment - New Animated Short Film I just watched this new animated short film and I gotta say, I'm left feeling a bit underwhelmed. Now, don't get me wrong, it's not all bad. The visuals are bright and colorful, and the animation is smooth and fluid. But something about it just didn't sit right wit...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	User Review: Entertainment - New Animated Short Film I just watched this new animated short film and I gotta say, I'm left feeling a bit underwhelmed. Now, don't get me wrong, it's not all bad. The visuals are bright and colorful, and the animation is smooth and fluid. But something about it just didn't sit right with ...	0.918919	0.990533	2,803	1	0.633731	0.244173	0.269233	0.256914	0.764857	12.1875	14.602866	2.618271	0.041769	0.115479	-0.803387	11.926954	2.544176	0.03317	0.087224	0.856398	4.474043	1.133811	0.069182	0.015723	-4.931704	4.135923	1.572513	0.028302	0.012579	1.680403	0.862431	0.853832	0.020767	3	0.99951	0	0.019188	3	0.999961	0.00628	3	0.999978	0.016398	3	0.999997
Mission: The tennessee chapter of the american academy of pediatrics is committed to being a leading voice, advocate and authority for the physical, mental and social welfare of infants, children and adolescents, as well as the pediatricians who care for them. Programs: The organization works with the bureau of tenncar...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nImprove this text by rewriting it. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the t...	Health	About (Org.)	The Tennessee Chapter of the American Academy of Pediatrics is dedicated to being a prominent voice, advocate, and authority for the overall health and well-being of infants, children, and adolescents, as well as the pediatricians who care for them. In pursuit of this mission, the organization collaborates with the Bu...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	The Tennessee Chapter of the American Academy of Pediatrics is dedicated to being a prominent voice, advocate, and authority for the overall health and well-being of infants, children, and adolescents, as well as the pediatricians who care for them. In pursuit of this mission, the organization collaborates with the Bur...	0.726457	0.857143	684	0.375325	0.15307	0.141162	0.128882	0.135066	0.556431	-4.125	11.871114	2.353198	0.060484	0.068548	-0.846513	9.01115	2.375661	0.044355	0.052419	1.297445	10.470218	2.2828	0.051661	0.062731	-0.521934	8.936068	2.225317	0.02952	0.055351	0.293715	0.819425	0.860965	0.008836	0	0.104953	0	0.003555	1	0.420338	0.001896	1	0.294947	0.001881	1	0.433836
JAGGAER’s Amenallah Reghimi and Michael Roesch recently partnered with Philip Ideson, founder and managing director at Art of Procurement, to bring procurement professionals a presentation on response tactics for the current global crisis. Our speakers received far more questions than they were able to answer at the ti...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nFormulate a set of strict, absolute constraints regarding format, word count, vocabulary restrictions, and syntax rules that, if followed perfectly by an AI, would result in generating a text nearly identical to this one. Note that your response should be for...	Finance & Business	Audio Transcript	Generate a document that is formatted exactly like the provided one, using the same headings and structure. The content should be focused on a Q&A session about supply chain management and procurement strategies, specifically addressing topics such as change management, market price updates, supplier relationships, ri...	Finance & Business: Supply Chain Management and Procurement Strategies Interviewer: In today's ever-changing business landscape, how can companies effectively manage changes in their supply chains? Expert: Change management in supply chain management is crucial for businesses to adapt and thrive. Companies s...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	Finance & Business: Supply Chain Management and Procurement Strategies Interviewer: In today's ever-changing business landscape, how can companies effectively manage changes in their supply chains? Expert: Change management in supply chain management is crucial for businesses to adapt and thrive. Companies shou...	0.891727	0.99192	6,143	0.990985	0.441778	0.250955	0.258369	0.254681	0.788485	2.5625	15.611532	2.658813	0.042937	0.091475	-1.666081	12.337936	2.525979	0.039826	0.074673	0.262874	4.023307	1.100151	0.064394	0.020833	-4.833157	3.704812	1.329561	0.030303	0.020833	0.305322	0.854275	0.936532	0.010544	3	0.999579	1	0.238806	3	0.999918	0.001734	3	0.997572	0.002784	3	0.995097
The Pack contains associated resources for the learning experience, typically in the form of articles and videos. There is a teacher Pack (with only teacher information) and a student Pack (which contains only student information). As a teacher, you can toggle between both to see everything. Here are the teacher pack i...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nWrite a prequel to this text, describing the events, thoughts, or historical context that immediately led up to this piece being written, making sure to use similar language and wording as the author, as if it was written by the author.\nOutput the full new t...	History	Product Page	Product Page: In the lead up to 1968, the United States found itself in the midst of significant social and political change. The country was grappling with issues such as civil rights, the Vietnam War, and poverty, leading to widespread unrest among many groups, including college students. As the year approached, te...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	Product Page: In the lead up to 1968, the United States found itself in the midst of significant social and political change. The country was grappling with issues such as civil rights, the Vietnam War, and poverty, leading to widespread unrest among many groups, including college students. As the year approached, tens...	0.789683	0.898551	1,276	0.788284	0.219056	0.197195	0.155158	0.176712	0.644261	-2	15.51147	2.525018	0.042918	0.11588	-1.487888	11.264628	2.504724	0.030043	0.107296	0.602275	5.487837	1.593121	0.045455	0.033333	-1.158065	5.244263	1.781942	0.021212	0.033333	1.308674	0.842289	0.883417	0.074456	1	0.42167	0	0.063779	3	0.98653	0.008064	3	0.998367	0.006961	3	0.926929
Ever thought of adding Barcelona’s greatest libraries to your must-see itinerary? In this post, we’ll tell you all about the best-kept secrets of Barcelona’s literary scene, with some of the most spectacular buildings to see and perfectly complete your visit to Barcelona. Biblioteca Jaume Foster Architect Josep Llinàs ...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nEnvision the scenario in which the author wrote this text. Describe that scenario in exhaustive detail.\nDo not output anything besides what you were requested to write, and do not output any extra commentary.", "{{RESP_0}}\nAbove is an AI generated descr...	Travel	Listicle	The author is sitting at a desk in a quiet room, surrounded by books and papers. There are photographs and postcards of Barcelona's libraries pinned to a corkboard above the desk. The author has been researching the topic, taking notes and gathering information about the libraries' histories, architecture, and collect...	Barcelona: A Literary Tour of the City's Hidden Gems Barcelona is a city steeped in history, culture, and beauty. While famous landmarks like La Sagrada Familia and Park Guell rightfully attract throngs of visitors, there is another side to Barcelona that is worth exploring: its libraries. These institutions, often o...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	Barcelona: A Literary Tour of the City's Hidden Gems Barcelona is a city steeped in history, culture, and beauty. While famous landmarks like La Sagrada Familia and Park Guell rightfully attract throngs of visitors, there is another side to Barcelona that is worth exploring: its libraries. These institutions, often ove...	0.882979	0.965744	2,396	0.925644	0.177451	0.19642	0.180693	0.188632	0.647824	1.0625	13.989013	2.610703	0.053279	0.094262	-0.277767	11.043947	2.446635	0.045082	0.088115	0.463752	3.469802	1.021627	0.046784	0.017544	-4.249848	3.138029	1.195714	0.020468	0.017544	0.941121	0.831959	0.839431	0.013881	3	0.999271	0	0.012652	3	0.999988	0.002722	3	0.99996	0.001667	3	0.999995
OVERLAND PARK, Kan. (WDAF) — Patrick Mahomes trading card values are soaring. According to Barrons.com, just last week a Mahomes signed rookie card sold for $861,000! A Kansas City collector has a small stash that might be able to top that. Scott Neal has been collecting since he was a kid. He owns The Baseball Card St...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nExpand this text by adding real-world applications and specific details.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you have written.", "...	Sports & Fitness	News Article	Overland Park, Kan. (WDAF) - The value of Patrick Mahomes trading cards has skyrocketed in recent times, reaching unprecedented levels. According to Barron's, a Mahomes signed rookie card was sold for a staggering $861,00	Overland Park, Kansas: A Patrick Mahomes signed rookie card was sold for an astonishing $861,000, according to Barron's. This sale showcases the soaring value of Mahomes' trading cards, which have reached unprecedented heights.	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	Overland Park, Kansas: A Patrick Mahomes signed rookie card was sold for an astonishing $861,000, according to Barron's. This sale showcases the soaring value of Mahomes' trading cards, which have reached unprecedented heights.	0.944606	0.990926	2,912	0.462857	0.159605	0.186142	0.290092	0.241663	0.785277	-7.5	16.442063	2.625354	0.047143	0.094286	-2.09255	12.172782	2.378123	0.051429	0.082857	-1.58295	18.848649	2.502024	0.125	0.145833	-1.35519	18.770197	2.997092	0.0625	0.145833	0.191651	0.836199	0.984179	0.024367	1	0.503507	0	0.028437	1	0.479814	0.002368	3	0.827494	0.002452	1	0.52108
IRON SEAL™ has been changing in recent weeks. The main changes concern the expansion of the known world and the relationship between characters. Read on to find out what JAMATORA plan to do with this new game set in the unusual world. When we started working on "Iron Seal", the world seemed simple. Just one continent s...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nRestructure this text to improve the transitions and connections between ideas.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you have written."...	Games	News (Org.)	When the team at JAMATORA began developing "Iron Seal," the world was initially conceived as a single continent surrounded by seas and oceans. While the map was visually appealing and made it easy for players to navigate, the simplicity of the world became apparent. To create more challenges for players, JAMATORA deci...	In the realm of game development, creators like Jamatora are constantly seeking ways to enhance the player's experience by introducing innovative elements and challenges. For their game "Iron Seal," the initial concept featured a single continent surrounded by vast seas, providing a simple and easily navigable world. ...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	In the realm of game development, creators like Jamatora are constantly seeking ways to enhance the player's experience by introducing innovative elements and challenges. For their game "Iron Seal," the initial concept featured a single continent surrounded by vast seas, providing a simple and easily navigable world. H...	0.888889	0.974093	2,106	0.951282	0.298475	0.235306	0.199025	0.217586	0.721387	1.75	24.548559	2.96882	0.049242	0.140152	-1.573548	19.96904	2.98391	0.037879	0.125	-0.071456	5.833658	1.513301	0.054393	0.037657	-3.404561	5.905954	1.842057	0.025105	0.039749	0.827167	0.925894	0.961479	0.012867	3	0.99962	0	0.011538	3	0.999996	0.002247	3	0.999928	0.006004	3	0.999562
Epoxy floor coating is a great choice for manufacturing plants, storage and warehousing flooring. Honestone’s epoxy coating is ideal for concrete floors in industrial and commercial environments. It ensures a smooth, durable and patch-free surface that can withstand heavy weight and large traffic. Our epoxy toppings ar...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nWrite a piece of text in the same style and language as this text, but is directly contrasting to the themes, ideas, and/or details in this text.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not...	Industrial	Product Page	Product Page: Epoxy-Free Flooring Solution - Honestone's Alternative to Epoxy Flooring --------------- At Honestone, we understand that epoxy flooring isn't always the perfect fit for every project. That's why we offer an epoxy-free flooring solution for those who seek alternative benefits and features. Our innovati...		mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	At Honestone, we understand that epoxy flooring isn't always the perfect fit for every project. That's why we offer an epoxy-free flooring solution for those who seek alternative benefits and features. Our innovative product provides a seamless, easy-to-clean surface that requires minimal maintenance compared to epoxy ...	0.908907	0.991736	2,621	0.829726	0.45776	0.240795	0.231575	0.236213	0.750332	5.125	15.693652	2.550047	0.053296	0.078541	-2.583653	12.443767	2.497524	0.047686	0.065919	-0.314192	4.617508	1.244595	0.049383	0.02716	-3.815216	4.650456	1.483772	0.02963	0.02963	-0.65504	0.905045	0.959064	0.011589	3	0.961016	0	0.003059	3	0.973733	0.001704	2	0.616314	0.000829	3	0.832133
Since I've had the nesting bug so bad I've been pinning like crazy! If you don't follow me yet on Pinterest you can follow here. Here are some things on my mind lately: 1. DIY Burlap Wreath I actually made this last night and it's much easier than I expected it to be! There was definitely a small learning curve but onc...	{ "chat_turns": [ "<document>\n{{DOC}}\n</document>\n\nClarify this text by paraphrasing it with different vocabulary and varied sentence structures.\nOutput the full new text with no extra statements or commentations.\nBegin directly with the text itself. Do not add a title, a heading, or a label naming what you...	Home & Hobbies	Listicle	I've been obsessed with creating nested designs recently, and if you're not already following me on Pinterest, you can find me here. Here are some ideas that have been on my mind lately: 1. Burlap Wreath I created this wreath last night, and it turned out to be simpler than I initially thought! Although there was a s...	I've been captivated by creating nested designs recently, and if you're not already following me on Pinterest, you can find me here. Here are some ideas that have been on my mind lately: 1. Burlap Wreath: After deciding to create this wreath last night, I discovered it was simpler than I initially thought! Despite a ...	mistralai/Mixtral-8x7B-Instruct-v0.1	{"temperature": 0.7, "top_p": 0.95, "extra_body": {"top_k": 40, "repetition_penalty": 1.1, "chat_template_kwargs": {"enable_thinking": false}}}	I've been captivated by creating nested designs recently, and if you're not already following me on Pinterest, you can find me here. Here are some ideas that have been on my mind lately: 1. Burlap Wreath: After deciding to create this wreath last night, I discovered it was simpler than I initially thought! Despite a sl...	0.775244	0.923554	977	0.74482	0.141775	0.156321	0.1511	0.153719	0.579656	-4.0625	16.575778	2.461175	0.059006	0.102484	-2.775696	11.371796	2.550407	0.034161	0.062112	1.004829	13.597126	2.381386	0.046921	0.102639	-2.051592	12.13263	2.502112	0.035191	0.082111	0.056274	0.820115	0.900828	0.028645	1	0.420286	0	0.017138	1	0.451577	0.011331	2	0.823398	0.004761	2	0.584488

End of preview. Expand in Data Studio

Auto-generated FastDetector dataset

Fastdetector Val Stat

Best detector

Giga EditLens Llama-3.2-3B Score

0.6638 TPR @ 1% FPR

Hardest prompt subset

rewrite

0.3529 max detector TPR @ 1% FPR

Hardest generator config

hy3 (Temp: Unknown)

0.5385 max detector TPR @ 1% FPR

13,555

rows

generator configs

prompt subsets

detectors

01Leaderboard 02Model analytics 03Distances 04Appendix

Detector leaderboard

Score-based detectors ranked by overall AUROC. Thresholds are placed exactly on every human score, so TPR is reported at a true 1% and 0.1% false positive rate. Open a row for its threshold sweep, its TPR as AI rows with low cosdist are dropped, and its score distributions.

01Giga EditLens Llama-3.2-3B ScoreN27,110AUROC0.9205TPR @ 1% FPR0.6638TPR @ 0.1% FPR0.4440⌄

1% FPR threshold 0.16600.1% FPR threshold 0.5000

Threshold sweep: Giga EditLens Llama-3.2-3B Score

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Giga EditLens Llama-3.2-3B Score: TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Giga EditLens Llama-3.2-3B Score: overall human

Overall · human onlyOpen full size ↗

Giga EditLens Llama-3.2-3B Score: overall AI

Overall · AI onlyOpen full size ↗

Giga EditLens Llama-3.2-3B Score: prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Giga EditLens Llama-3.2-3B Score: generator subsets

AI texts · one envelope per generator configOpen full size ↗

02EditLens Llama-3.2-3B ScoreN27,110AUROC0.9047TPR @ 1% FPR0.6293TPR @ 0.1% FPR0.4450⌄

1% FPR threshold 0.30290.1% FPR threshold 0.5154

Threshold sweep: EditLens Llama-3.2-3B Score

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

EditLens Llama-3.2-3B Score: TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

EditLens Llama-3.2-3B Score: overall human

Overall · human onlyOpen full size ↗

Overall · AI onlyOpen full size ↗

EditLens Llama-3.2-3B Score: prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

EditLens Llama-3.2-3B Score: generator subsets

AI texts · one envelope per generator configOpen full size ↗

03Giga EditLens Roberta ScoreN27,110AUROC0.8948TPR @ 1% FPR0.5273TPR @ 0.1% FPR0.2331⌄

1% FPR threshold 0.51340.1% FPR threshold 0.9948

Threshold sweep: Giga EditLens Roberta Score

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Giga EditLens Roberta Score: TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Giga EditLens Roberta Score: overall human

Overall · human onlyOpen full size ↗

Overall · AI onlyOpen full size ↗

Giga EditLens Roberta Score: prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Giga EditLens Roberta Score: generator subsets

AI texts · one envelope per generator configOpen full size ↗

04EditLens Roberta-Large ScoreN27,110AUROC0.8762TPR @ 1% FPR0.4361TPR @ 0.1% FPR0.2229⌄

1% FPR threshold 0.46170.1% FPR threshold 0.9670

Threshold sweep: EditLens Roberta-Large Score

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

EditLens Roberta-Large Score: TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

EditLens Roberta-Large Score: overall human

Overall · human onlyOpen full size ↗

EditLens Roberta-Large Score: overall AI

Overall · AI onlyOpen full size ↗

EditLens Roberta-Large Score: prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

EditLens Roberta-Large Score: generator subsets

AI texts · one envelope per generator configOpen full size ↗

05Perplexity (Llama-3.2-3B-Instruct)N27,110AUROC0.6455TPR @ 1% FPR0.1382TPR @ 0.1% FPR0.0185⌄

1% FPR threshold 6.18030.1% FPR threshold 3.0318

Threshold sweep: Perplexity (Llama-3.2-3B-Instruct)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Perplexity (Llama-3.2-3B-Instruct): overall human

Overall · human onlyOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): overall AI

Overall · AI onlyOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): generator subsets

AI texts · one envelope per generator configOpen full size ↗

06Entropy (Llama-3.2-3B-Instruct)N27,110AUROC0.6421TPR @ 1% FPR0.1318TPR @ 0.1% FPR0.0244⌄

1% FPR threshold 1.62970.1% FPR threshold 1.0552

Threshold sweep: Entropy (Llama-3.2-3B-Instruct)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Entropy (Llama-3.2-3B-Instruct): overall human

Overall · human onlyOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): overall AI

Overall · AI onlyOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): generator subsets

AI texts · one envelope per generator configOpen full size ↗

07Top-k Outliers (Llama-3.2-3B-Instruct)N27,110AUROC0.6202TPR @ 1% FPR0.1065TPR @ 0.1% FPR0.0261⌄

1% FPR threshold 0.03140.1% FPR threshold 0.0124

Threshold sweep: Top-k Outliers (Llama-3.2-3B-Instruct)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Top-k Outliers (Llama-3.2-3B-Instruct): overall human

Overall · human onlyOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): overall AI

Overall · AI onlyOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): generator subsets

AI texts · one envelope per generator configOpen full size ↗

08BinocularsN27,110AUROC0.5967TPR @ 1% FPR0.0257TPR @ 0.1% FPR0.0038⌄

1% FPR threshold 1.00910.1% FPR threshold 1.0754

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Overall · human onlyOpen full size ↗

Overall · AI onlyOpen full size ↗

AI texts · one envelope per prompt subsetOpen full size ↗

AI texts · one envelope per generator configOpen full size ↗

09Perplexity (Llama-3.2-3B)N27,110AUROC0.5959TPR @ 1% FPR0.0249TPR @ 0.1% FPR0.0001⌄

1% FPR threshold 3.25740.1% FPR threshold 1.2173

Threshold sweep: Perplexity (Llama-3.2-3B)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Perplexity (Llama-3.2-3B): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Perplexity (Llama-3.2-3B): overall human

Overall · human onlyOpen full size ↗

Overall · AI onlyOpen full size ↗

Perplexity (Llama-3.2-3B): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Perplexity (Llama-3.2-3B): generator subsets

AI texts · one envelope per generator configOpen full size ↗

10Top-p Outliers (Llama-3.2-3B)N27,110AUROC0.5946TPR @ 1% FPR0.0302TPR @ 0.1% FPR0.0000⌄

1% FPR threshold 0.01170.1% FPR threshold -0.0000

Threshold sweep: Top-p Outliers (Llama-3.2-3B)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Top-p Outliers (Llama-3.2-3B): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Top-p Outliers (Llama-3.2-3B): overall human

Overall · human onlyOpen full size ↗

Top-p Outliers (Llama-3.2-3B): overall AI

Overall · AI onlyOpen full size ↗

Top-p Outliers (Llama-3.2-3B): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Top-p Outliers (Llama-3.2-3B): generator subsets

AI texts · one envelope per generator configOpen full size ↗

11Entropy (Llama-3.2-3B)N27,110AUROC0.5896TPR @ 1% FPR0.0109TPR @ 0.1% FPR0.0002⌄

1% FPR threshold 1.16460.1% FPR threshold 0.2281

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Entropy (Llama-3.2-3B): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Overall · human onlyOpen full size ↗

Overall · AI onlyOpen full size ↗

AI texts · one envelope per prompt subsetOpen full size ↗

Entropy (Llama-3.2-3B): generator subsets

AI texts · one envelope per generator configOpen full size ↗

12Top-k Outliers (Llama-3.2-3B)N27,110AUROC0.5790TPR @ 1% FPR0.0443TPR @ 0.1% FPR0.0018⌄

1% FPR threshold 0.01670.1% FPR threshold 0.0012

Threshold sweep: Top-k Outliers (Llama-3.2-3B)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Top-k Outliers (Llama-3.2-3B): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Top-k Outliers (Llama-3.2-3B): overall human

Overall · human onlyOpen full size ↗

Top-k Outliers (Llama-3.2-3B): overall AI

Overall · AI onlyOpen full size ↗

Top-k Outliers (Llama-3.2-3B): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Top-k Outliers (Llama-3.2-3B): generator subsets

AI texts · one envelope per generator configOpen full size ↗

13Top-p Outliers (Llama-3.2-3B-Instruct)N27,110AUROC0.5417TPR @ 1% FPR0.0347TPR @ 0.1% FPR0.0043⌄

1% FPR threshold 0.02810.1% FPR threshold 0.0136

Threshold sweep: Top-p Outliers (Llama-3.2-3B-Instruct)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

Top-p Outliers (Llama-3.2-3B-Instruct): overall human

Overall · human onlyOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): overall AI

Overall · AI onlyOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): generator subsets

AI texts · one envelope per generator configOpen full size ↗

14FastDetectGPT (Llama-3.2-3B-Instruct)N27,110AUROC0.5104TPR @ 1% FPR0.0097TPR @ 0.1% FPR0.0007⌄

1% FPR threshold 2.56830.1% FPR threshold 5.6451

Threshold sweep: FastDetectGPT (Llama-3.2-3B-Instruct)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

FastDetectGPT (Llama-3.2-3B-Instruct): overall human

Overall · human onlyOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): overall AI

Overall · AI onlyOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): generator subsets

AI texts · one envelope per generator configOpen full size ↗

15FastDetectGPT (Llama-3.2-3B)N27,110AUROC0.4356TPR @ 1% FPR0.0193TPR @ 0.1% FPR0.0033⌄

1% FPR threshold -2.71390.1% FPR threshold -3.8349

Threshold sweep: FastDetectGPT (Llama-3.2-3B)

TPR and FPR at every threshold · both FPR operating points markedOpen full size ↗

FastDetectGPT (Llama-3.2-3B): TPR against minimum cosdist

AI rows with cosdist below the cutoff are dropped · human rows and thresholds are unchangedOpen full size ↗

Score distributions

FastDetectGPT (Llama-3.2-3B): overall human

Overall · human onlyOpen full size ↗

FastDetectGPT (Llama-3.2-3B): overall AI

Overall · AI onlyOpen full size ↗

FastDetectGPT (Llama-3.2-3B): prompt subsets

AI texts · one envelope per prompt subsetOpen full size ↗

FastDetectGPT (Llama-3.2-3B): generator subsets

AI texts · one envelope per generator configOpen full size ↗

Model-Specific Analytics

Select a model, then a split; the split's charts appear below it. Topic × format compares human scores, AI scores and their unsigned gap; generator × prompt category shows only the AI side, since neither changes the human text; specific prompt ranks every first message by difficulty.

Select a model

01Giga EditLens Llama-3.2-3B ScoreAUROC 0.9205 · TPR@1% 0.6638⌄

Select a split · Giga EditLens Llama-3.2-3B Score

Topic × formathuman · AI · |AI − human| gridsshow ▾

Giga EditLens Llama-3.2-3B Score: Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Giga EditLens Llama-3.2-3B Score: AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Giga EditLens Llama-3.2-3B Score: Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Giga EditLens Llama-3.2-3B Score: AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

One row per distinct first message (a second turn, where there is one, is ignored), hardest first: ranked by TPR at the detector's corpus-wide 1% FPR threshold. Score columns are the detector's mean over the prompt's rows, |AI − human| is the mean per-pair score movement, and distance columns are the mean distance between each source and its rewrite. Only the 12 hardest of 361 are shown; the full ranking ↗ is a CSV beside this card.

Prompt (first message)	Rows	AUROC	TPR @ 1% FPR	TPR @ 0.1% FPR	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	12	0.7847	0.0000	0.0000	0.0160	0.0481	0.0350	0.1072	0.0279
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	23	0.6011	0.0000	0.0000	0.0103	0.0155	0.0063	0.2011	0.0363
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.7041	0.0000	0.0000	0.0064	0.0175	0.0135	0.2155	0.0375
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	15	0.5733	0.0000	0.0000	0.0194	0.0173	0.0120	0.1433	0.0126
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6154	0.0000	0.0000	0.0143	0.0194	0.0068	0.1918	0.0128
Restructure this text to arrange the points in order of importance. The final text must have the topic <<topic>>. revise · single-turn	5	0.6800	0.0000	0.0000	0.0137	0.0071	0.0133	0.1480	0.0333
Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the format <<format>>. rewrite · single-turn	15	0.7667	0.0667	0.0000	0.0066	0.0479	0.0413	0.1320	0.0289
Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	15	0.7067	0.0667	0.0000	0.0139	0.0471	0.0334	0.1236	0.0183
Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	14	0.6531	0.0714	0.0714	0.0078	0.0705	0.0634	0.1797	0.0359
Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched. The final text must have the format <<format>>. rewrite · single-turn	13	0.6923	0.0769	0.0000	0.0123	0.0343	0.0220	0.2582	0.0457
Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6154	0.0769	0.0000	0.0053	0.0265	0.0217	0.1403	0.0027
Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	12	0.7361	0.0833	0.0000	0.0213	0.0544	0.0331	0.1312	0.0254

02EditLens Llama-3.2-3B ScoreAUROC 0.9047 · TPR@1% 0.6293⌄

Select a split · EditLens Llama-3.2-3B Score

Topic × formathuman · AI · |AI − human| gridsshow ▾

EditLens Llama-3.2-3B Score: Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

EditLens Llama-3.2-3B Score: AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

EditLens Llama-3.2-3B Score: Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

EditLens Llama-3.2-3B Score: AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched. The final text must have the format <<format>>. rewrite · single-turn	13	0.6746	0.0580	0.0830	0.0277	0.2582	0.0457
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	12	0.7083	0.0430	0.0990	0.0562	0.1072	0.0279
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the topic <<topic>>. rewrite · single-turn	6	0.7500	0.0201	0.0555	0.0354	0.0958	0.0225
Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	12	0.7361	0.0423	0.0938	0.0517	0.1312	0.0254
Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6154	0.0261	0.0493	0.0341	0.1403	0.0027
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	23	0.5369	0.0335	0.0404	0.0153	0.2011	0.0363
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6805	0.0310	0.0425	0.0333	0.2155	0.0375
Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the format <<format>>. rewrite · single-turn	15	0.6867	0.0361	0.0867	0.0553	0.1320	0.0289
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	15	0.5244	0.0694	0.0654	0.0346	0.1433	0.0126
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>>. rewrite · single-turn	14	0.6684	0.0367	0.0750	0.0405	0.1468	0.0139
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6036	0.0630	0.0863	0.0277	0.1918	0.0128
Restructure this text to arrange the points in order of importance. The final text must have the topic <<topic>>. revise · single-turn	5	0.4400	0.1106	0.0666	0.0688	0.1480	0.0333

03Giga EditLens Roberta ScoreAUROC 0.8948 · TPR@1% 0.5273⌄

Select a split · Giga EditLens Roberta Score

Topic × formathuman · AI · |AI − human| gridsshow ▾

Giga EditLens Roberta Score: Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Giga EditLens Roberta Score: AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Giga EditLens Roberta Score: Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Giga EditLens Roberta Score: AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	12	0.8333	0.0080	0.0706	0.0640	0.1072	0.0279
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>>. rewrite · single-turn	16	0.7109	0.0543	0.1660	0.1190	0.2296	0.0513
Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the format <<format>>. rewrite · single-turn	11	0.7438	0.0047	0.1116	0.1073	0.2336	0.0242
Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.5503	0.0052	0.0300	0.0255	0.1403	0.0027
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.6124	0.0057	0.0107	0.0072	0.2155	0.0375
Edit this text by replacing all instances of passive voice with active voice, without altering any sentences that are already in the active voice. The final text must have the format <<format>>. rewrite · single-turn	20	0.6650	0.0120	0.0418	0.0352	0.1619	0.0147
Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the format <<format>>. rewrite · single-turn	15	0.7889	0.0155	0.0894	0.0741	0.1320	0.0289
Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	15	0.5622	0.0336	0.0532	0.0197	0.1236	0.0183
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	15	0.5644	0.0152	0.0223	0.0093	0.1433	0.0126
Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.5030	0.0439	0.0528	0.0216	0.1918	0.0128
Restructure this text to arrange the points in order of importance. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	4	0.5000	0.0065	0.1092	0.1031	0.2627	0.0491
Restructure this text to arrange the points in order of importance. The final text must have the topic <<topic>>. revise · single-turn	5	0.7600	0.0059	0.0095	0.0061	0.1480	0.0333

04EditLens Roberta-Large ScoreAUROC 0.8762 · TPR@1% 0.4361⌄

Select a split · EditLens Roberta-Large Score

Topic × formathuman · AI · |AI − human| gridsshow ▾

EditLens Roberta-Large Score: Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

EditLens Roberta-Large Score: AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

EditLens Roberta-Large Score: Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

EditLens Roberta-Large Score: AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Clarify this text by improving the phrasing while preserving its original intent. The final text must have the format <<format>>. revise · single-turn	5	1.0000	0.0144	0.3113	0.2969	0.5350	0.0346
Edit this text by completely rewriting the topic sentence of every paragraph, while preserving the supporting sentences in those paragraphs verbatim. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	19	0.7341	0.0248	0.0974	0.0727	0.2413	0.0449
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	12	0.7431	0.0312	0.0861	0.0548	0.1072	0.0279
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>>. rewrite · single-turn	16	0.6094	0.0876	0.1222	0.0709	0.2296	0.0513
Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the topic <<topic>>. rewrite · single-turn	6	0.6389	0.0536	0.0550	0.0148	0.0958	0.0225
Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	12	0.6528	0.0440	0.1021	0.0625	0.1312	0.0254
Edit this text by inserting twenty to thirty words (in appropriate locations), without changing anything else. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	22	0.7521	0.0400	0.1161	0.0798	0.2616	0.0469
Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.7101	0.0161	0.0365	0.0215	0.1403	0.0027
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	23	0.5435	0.1056	0.1158	0.0502	0.2011	0.0363
Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the topic <<topic>>. rewrite · single-turn	13	0.5533	0.0436	0.0359	0.0292	0.2155	0.0375
Edit this text by removing five to ten percent of the words from it (the least necessary), without changing anything else. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	19	0.7452	0.0503	0.1267	0.0850	0.3222	0.0370
Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow. The final text must have the format <<format>> and topic <<topic>>. rewrite · single-turn	14	0.6097	0.0294	0.0566	0.0311	0.1797	0.0359

05Perplexity (Llama-3.2-3B-Instruct)AUROC 0.6455 · TPR@1% 0.1382⌄

Select a split · Perplexity (Llama-3.2-3B-Instruct)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Perplexity (Llama-3.2-3B-Instruct): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Perplexity (Llama-3.2-3B-Instruct): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Perplexity (Llama-3.2-3B-Instruct): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.6627	21.5173	14.9724	8.0329	0.7688	0.1523
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.8062	22.7118	13.3853	9.9076	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.8750	15.2299	10.2386	4.9913	0.7989	0.1939
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.9844	27.3579	12.1629	15.1950	0.6826	0.1469
Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	9	0.5926	19.1183	17.6020	7.7920	0.8113	0.1859
Adapt this text for social media sharing. The final text must have the format <<format>>. revise · single-turn	6	0.6389	17.7672	14.6084	3.1588	0.5916	0.1495
Adapt this text for social media sharing. The final text must have the topic <<topic>>. revise · single-turn	7	0.3061	15.4253	18.8433	8.3449	0.7882	0.2264
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	10	0.5200	17.4573	21.5327	9.7465	0.7588	0.0719
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>>. revise · single-turn	10	0.7000	14.9120	12.4268	4.1531	0.7524	0.1003
Change the tone of this text to be more academic and scholarly. The final text must have the topic <<topic>>. revise · single-turn	13	0.7160	16.9389	11.9811	6.3284	0.7067	0.1562
Change the tone of this text to be more analytical and logical. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	14	0.4898	19.2931	19.5711	7.8765	0.8208	0.1699
Change the tone of this text to be more analytical and logical. The final text must have the topic <<topic>>. revise · single-turn	7	0.5510	15.8698	15.1792	8.6561	0.7783	0.1760

06Entropy (Llama-3.2-3B-Instruct)AUROC 0.6421 · TPR@1% 0.1318⌄

Select a split · Entropy (Llama-3.2-3B-Instruct)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Entropy (Llama-3.2-3B-Instruct): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Entropy (Llama-3.2-3B-Instruct): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Entropy (Llama-3.2-3B-Instruct): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.8512	2.7832	2.2335	0.5750	0.6964	0.1311
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.9375	3.0568	2.3195	0.7614	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the topic <<topic>>. revise · single-turn	11	0.6364	2.5651	2.3919	0.5815	0.7994	0.2490
Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	9	0.5926	2.5394	2.2774	0.4314	0.8113	0.1859
Adapt this text for social media sharing. The final text must have the topic <<topic>>. revise · single-turn	7	0.6735	2.4237	2.3161	0.3833	0.7882	0.2264
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	10	0.4200	2.5610	2.7324	0.3513	0.7588	0.0719
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>>. revise · single-turn	10	0.5600	2.3620	2.3338	0.2480	0.7524	0.1003
Change the tone of this text to be more academic and scholarly. The final text must have the topic <<topic>>. revise · single-turn	13	0.6331	2.5302	2.3701	0.3708	0.7067	0.1562
Change the tone of this text to be more analytical and logical. revise · single & two-turn	54	0.5888	2.7407	2.5678	0.4558	0.7541	0.1475
Change the tone of this text to be more analytical and logical. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	14	0.4694	2.6488	2.6402	0.3306	0.8208	0.1699
Change the tone of this text to be more conversational and approachable. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7969	2.6661	2.2346	0.4315	0.7456	0.0818
Change the tone of this text to be more conversational and approachable. The final text must have the topic <<topic>>. revise · single-turn	9	0.7407	2.7546	2.1957	0.8853	0.7736	0.1301

07Top-k Outliers (Llama-3.2-3B-Instruct)AUROC 0.6202 · TPR@1% 0.1065⌄

Select a split · Top-k Outliers (Llama-3.2-3B-Instruct)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Top-k Outliers (Llama-3.2-3B-Instruct): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Top-k Outliers (Llama-3.2-3B-Instruct): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Top-k Outliers (Llama-3.2-3B-Instruct): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7969	0.0958	0.0611	0.0408	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.7924	0.1214	0.0762	0.0494	0.6964	0.1311
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	1.0000	0.1460	0.0735	0.0725	0.6826	0.1469
Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	9	0.5185	0.1026	0.1050	0.0380	0.8113	0.1859
Adapt this text for social media sharing. The final text must have the format <<format>>. revise · single-turn	6	0.6389	0.1045	0.0951	0.0248	0.5916	0.1495
Adapt this text for social media sharing. The final text must have the topic <<topic>>. revise · single-turn	7	0.1837	0.0948	0.1225	0.0491	0.7882	0.2264
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	10	0.4500	0.1063	0.1224	0.0367	0.7588	0.0719
Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>>. revise · single-turn	10	0.5500	0.0933	0.0942	0.0266	0.7524	0.1003
Change the tone of this text to be more academic and scholarly. The final text must have the topic <<topic>>. revise · single-turn	13	0.7396	0.1108	0.0798	0.0361	0.7067	0.1562
Change the tone of this text to be more analytical and logical. revise · single & two-turn	54	0.5806	0.1158	0.1043	0.0403	0.7541	0.1475
Change the tone of this text to be more analytical and logical. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	14	0.4031	0.1061	0.1139	0.0368	0.8208	0.1699
Change the tone of this text to be more analytical and logical. The final text must have the format <<format>>. revise · single-turn	11	0.5537	0.0977	0.0850	0.0463	0.7886	0.1486

08BinocularsAUROC 0.5967 · TPR@1% 0.0257⌄

Select a split · Binoculars

Topic × formathuman · AI · |AI − human| gridsshow ▾

Binoculars: Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Binoculars: AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Binoculars: Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Binoculars: AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.6875	0.7991	0.8944	0.1205	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.7500	0.8471	0.8948	0.0604	0.7848	0.1302
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.6406	0.8725	0.8945	0.0394	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.7197	0.8117	0.8984	0.0939	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.8125	0.8407	0.8726	0.0319	0.7989	0.1939
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.4219	0.8755	0.8689	0.0382	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.7602	0.8614	0.9067	0.0541	0.7535	0.1244
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7344	0.8728	0.8958	0.0451	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.6790	0.8541	0.8933	0.0659	0.7891	0.2360
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the topic <<topic>>. revise · single-turn	11	0.7603	0.8501	0.9028	0.0707	0.7994	0.2490
Adapt this text for social media sharing. The final text must have the format <<format>>. revise · single-turn	6	0.5278	0.8733	0.8933	0.0389	0.5916	0.1495
Change the tone of this text to be more academic and scholarly. revise · single & two-turn	50	0.5508	0.8504	0.8754	0.0600	0.8063	0.1711

09Perplexity (Llama-3.2-3B)AUROC 0.5959 · TPR@1% 0.0249⌄

Select a split · Perplexity (Llama-3.2-3B)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Perplexity (Llama-3.2-3B): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Perplexity (Llama-3.2-3B): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Perplexity (Llama-3.2-3B): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Perplexity (Llama-3.2-3B): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.6036	16.1795	13.0523	5.6131	0.7688	0.1523
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.6458	12.7609	10.6819	4.3834	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.8200	13.8999	9.0005	5.6786	0.7848	0.1302
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7969	12.6383	8.8605	3.8215	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.7405	15.4980	10.7884	6.2094	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.7500	11.5230	8.5273	2.9957	0.7989	0.1939
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.8125	13.2783	8.1820	5.0963	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.7092	14.1806	10.6204	5.5771	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.6281	11.1641	9.4208	3.2813	0.6904	0.0985
Adapt this text for an elderly adult with limited knowledge of modern day affairs. revise · single & two-turn	48	0.5911	16.0108	13.2109	7.1078	0.7629	0.1924
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.9688	20.7013	10.7336	9.9677	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.7037	17.7016	9.4230	9.4405	0.7891	0.2360

10Top-p Outliers (Llama-3.2-3B)AUROC 0.5946 · TPR@1% 0.0302⌄

Select a split · Top-p Outliers (Llama-3.2-3B)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Top-p Outliers (Llama-3.2-3B): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Top-p Outliers (Llama-3.2-3B): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Top-p Outliers (Llama-3.2-3B): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Top-p Outliers (Llama-3.2-3B): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.6597	0.0433	0.0346	0.0176	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.7450	0.0410	0.0288	0.0149	0.7848	0.1302
Adapt this text for a lay audience with no technical background. revise · single & two-turn	57	0.5996	0.0413	0.0369	0.0140	0.7827	0.1645
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7656	0.0449	0.0352	0.0150	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.6125	0.0381	0.0339	0.0148	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.8750	0.0419	0.0307	0.0132	0.7989	0.1939
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.9688	0.0435	0.0256	0.0183	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.6837	0.0473	0.0427	0.0140	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.5992	0.0407	0.0332	0.0129	0.6904	0.0985
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.6875	0.0463	0.0392	0.0107	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.5926	0.0419	0.0400	0.0125	0.7891	0.2360
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the topic <<topic>>. revise · single-turn	11	0.6364	0.0423	0.0374	0.0164	0.7994	0.2490

11Entropy (Llama-3.2-3B)AUROC 0.5896 · TPR@1% 0.0109⌄

Select a split · Entropy (Llama-3.2-3B)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Entropy (Llama-3.2-3B): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Entropy (Llama-3.2-3B): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Entropy (Llama-3.2-3B): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Entropy (Llama-3.2-3B): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.5266	2.5672	2.4937	0.3328	0.7688	0.1523
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.5417	2.2541	2.2815	0.4770	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.6700	2.6129	2.3352	0.3543	0.7848	0.1302
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.7969	2.4827	2.1361	0.3466	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.7163	2.5519	2.3741	0.4171	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.6875	2.4155	2.1668	0.2488	0.7989	0.1939
Adapt this text for a teenage audience. revise · single & two-turn	68	0.6386	2.4839	2.2740	0.3607	0.7178	0.1372
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.6719	2.4963	2.2029	0.3186	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.7143	2.5372	2.2346	0.4332	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.5207	2.3577	2.2742	0.2906	0.6904	0.0985
Adapt this text for an elderly adult with limited knowledge of modern day affairs. revise · single & two-turn	48	0.5859	2.5353	2.4423	0.3740	0.7629	0.1924
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.9062	2.9454	2.4457	0.5212	0.6826	0.1469

12Top-k Outliers (Llama-3.2-3B)AUROC 0.5790 · TPR@1% 0.0443⌄

Select a split · Top-k Outliers (Llama-3.2-3B)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Top-k Outliers (Llama-3.2-3B): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Top-k Outliers (Llama-3.2-3B): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Top-k Outliers (Llama-3.2-3B): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Top-k Outliers (Llama-3.2-3B): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.6568	0.1051	0.0775	0.0392	0.7688	0.1523
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.9000	0.0862	0.0448	0.0436	0.7848	0.1302
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.8594	0.0843	0.0524	0.0358	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.6990	0.0855	0.0597	0.0311	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.6875	0.0814	0.0578	0.0301	0.7989	0.1939
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.7041	0.0843	0.0582	0.0321	0.7535	0.1244
Adapt this text for an elderly adult with limited knowledge of modern day affairs. revise · single & two-turn	48	0.6107	0.0917	0.0794	0.0367	0.7629	0.1924
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.8594	0.1197	0.0646	0.0579	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.7778	0.0923	0.0525	0.0443	0.7891	0.2360
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the topic <<topic>>. revise · single-turn	11	0.6860	0.0735	0.0712	0.0463	0.7994	0.2490
Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	9	0.4815	0.0838	0.0903	0.0384	0.8113	0.1859
Adapt this text for social media sharing. The final text must have the format <<format>>. revise · single-turn	6	0.5000	0.0859	0.0920	0.0231	0.5916	0.1495

13Top-p Outliers (Llama-3.2-3B-Instruct)AUROC 0.5417 · TPR@1% 0.0347⌄

Select a split · Top-p Outliers (Llama-3.2-3B-Instruct)

Topic × formathuman · AI · |AI − human| gridsshow ▾

Top-p Outliers (Llama-3.2-3B-Instruct): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

Top-p Outliers (Llama-3.2-3B-Instruct): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

Top-p Outliers (Llama-3.2-3B-Instruct): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.6597	0.0622	0.0543	0.0148	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.4000	0.0491	0.0546	0.0108	0.7848	0.1302
Adapt this text for a lay audience with no technical background. revise · single & two-turn	57	0.4615	0.0584	0.0579	0.0155	0.7827	0.1645
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.5469	0.0605	0.0544	0.0134	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.4187	0.0586	0.0648	0.0157	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.5000	0.0550	0.0558	0.0046	0.7989	0.1939
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.3827	0.0578	0.0689	0.0190	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.4132	0.0563	0.0611	0.0101	0.6904	0.0985
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.6719	0.0618	0.0561	0.0152	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.5000	0.0528	0.0561	0.0128	0.7891	0.2360
Adapt this text for social media sharing. revise · single & two-turn	63	0.3584	0.0600	0.0694	0.0222	0.7993	0.2058
Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	9	0.3457	0.0602	0.0754	0.0278	0.8113	0.1859

14FastDetectGPT (Llama-3.2-3B-Instruct)AUROC 0.5104 · TPR@1% 0.0097⌄

Select a split · FastDetectGPT (Llama-3.2-3B-Instruct)

Topic × formathuman · AI · |AI − human| gridsshow ▾

FastDetectGPT (Llama-3.2-3B-Instruct): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

FastDetectGPT (Llama-3.2-3B-Instruct): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

FastDetectGPT (Llama-3.2-3B-Instruct): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. revise · single & two-turn	63	0.4402	-1.4960	-1.8380	1.3942	0.7791	0.1610
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.4852	-2.3823	-2.1979	1.5047	0.7688	0.1523
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.6111	-2.1263	-1.6294	1.3232	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.3300	-0.5625	-1.5381	1.7378	0.7848	0.1302
Adapt this text for a lay audience with no technical background. revise · single & two-turn	57	0.4097	-1.3601	-1.8871	1.3340	0.7827	0.1645
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.5781	-2.0228	-1.8367	1.6559	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.3979	-1.8742	-2.4876	1.1370	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.3750	-2.0811	-2.5502	1.0189	0.7989	0.1939
Adapt this text for a teenage audience. revise · single & two-turn	68	0.4872	-1.8662	-1.8193	1.3494	0.7178	0.1372
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.5781	-2.1183	-1.5657	1.3922	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.3265	-1.4954	-2.2075	1.4138	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.2397	-1.5248	-2.7839	1.6527	0.6904	0.0985

15FastDetectGPT (Llama-3.2-3B)AUROC 0.4356 · TPR@1% 0.0193⌄

Select a split · FastDetectGPT (Llama-3.2-3B)

Topic × formathuman · AI · |AI − human| gridsshow ▾

FastDetectGPT (Llama-3.2-3B): Human texts · mean score per topic and format

Human texts · mean score per topic and formatOpen full size ↗

FastDetectGPT (Llama-3.2-3B): AI texts · mean score, same colour scale as the human grid

AI texts · mean score, same colour scale as the human gridOpen full size ↗

FastDetectGPT (Llama-3.2-3B): Mean unsigned |AI − human| score per pair

Mean unsigned |AI − human| score per pairOpen full size ↗

Generator × prompt categoryAI · |AI − human| gridsshow ▾

FastDetectGPT (Llama-3.2-3B): AI texts · mean score per generator config and prompt category

AI texts · mean score per generator config and prompt categoryOpen full size ↗

Mean unsigned |AI − human| score per pairOpen full size ↗

Specific prompt361 prompts · hardest firstshow ▾

Prompt (first message)	Rows	AUROC	Human score	AI score	\|AI − human\|	jaccard_1	cosdist
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	13	0.2840	-0.5857	0.5081	1.2840	0.7688	0.1523
Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>. revise · single-turn	12	0.2639	-0.2835	0.6780	1.4834	0.7843	0.2194
Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>. revise · single-turn	10	0.1700	0.1705	1.5885	1.6571	0.7848	0.1302
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.3281	-0.0196	0.6533	1.2603	0.7070	0.1504
Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>. revise · single-turn	17	0.3183	-0.1966	0.4381	0.8569	0.6964	0.1311
Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>. revise · single-turn	4	0.1875	-0.3228	0.8379	1.3634	0.7989	0.1939
Adapt this text for a teenage audience. revise · single & two-turn	68	0.3692	-0.1848	0.4218	1.3197	0.7178	0.1372
Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.0312	-0.6112	2.1168	2.7280	0.7642	0.2463
Adapt this text for a teenage audience. The final text must have the format <<format>>. revise · single-turn	14	0.4541	-0.1575	0.0769	0.6169	0.7535	0.1244
Adapt this text for a teenage audience. The final text must have the topic <<topic>>. revise · single-turn	11	0.3802	0.0396	0.6879	0.9793	0.6904	0.0985
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>. revise · single-turn	8	0.1250	-0.3915	0.9078	1.2993	0.6826	0.1469
Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>. revise · single-turn	9	0.4074	0.0151	0.5048	1.6197	0.7891	0.2360

How far do the rewrites move?

Every distance measure between each source and its rewrite: overall, then with every prompt subset and every generator config overlaid on one chart each.

jaccard_1overall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

jaccard_2overall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

levenshteinoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

softngramoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

cosdistoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

bertscoreoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

bertscore_precisionoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

bertscore_recalloverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

moverscoreoverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

rerankeroverall · prompt subsets · generator configs⌄

OverallOpen full size ↗

One envelope per prompt subsetOpen full size ↗

One envelope per generator configOpen full size ↗

Appendix

Reference material: how the dataset was built and evaluated, its schema and loading code, univariate statistics, and cross-statistic correlations.

How the dataset was built7 generator configs · 4 prompt subsets · evaluation protocol⌄

Every row pairs original (Human) and final_response (AI): a human-written source and the text a generator model produced from it. Inclusion filtering happens while the dataset is built, so this report evaluates the published corpus exactly as it is.

Generator configs

Generator model	Temp	Sampling	Rows	Share
nvidia/Llama-4-Scout-17B-16E-Instruct-NVFP4	0.6	top_p 0.9 · enable_thinking False	1,944	14.3%
mistralai/Mixtral-8x7B-Instruct-v0.1	0.7	top_p 0.95 · top_k 40 · repetition_penalty 1.1 · enable_thinking False	1,959	14.5%
claude-sonnet-4-5-20250929	—	type disabled	1,935	14.3%
claude-sonnet-5	—	type disabled	1,964	14.5%
gpt-3.5-turbo	—	—	1,856	13.7%
gpt-4o	—	—	1,975	14.6%
tencent/hy3	—	enabled False · max_completion_tokens 16000	1,922	14.2%
Total			13,555	100%

Prompt subsets

Instructions are grouped by their first message. Counts cover all 13,555 rows.

direct_reference3,454 rows

single-turn · 60 instructions

Show the 60 instructions ▾

77× Write a new piece of text that is on a completely different topic, but reuses verbatim as many phrases and sentences from this text as possible. The final text must have the topic <<topic>>.
71× Write a different piece of text on the same topic. The final text must have the format <<format>>.
71× Write a followup to this piece of text, in the same style and language as the text. The final text must have the format <<format>> and topic <<topic>>.
71× Write another piece of text that is of a similar theme or background as this text. The final text must have the format <<format>>.
70× Write another piece of text that is of a similar theme or background as this text. The final text must have the format <<format>> and topic <<topic>>.
69× Write a different piece of text on the same topic. The final text must have the format <<format>> and topic <<topic>>.
68× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should use varied sentence lengths, and remove all common AI buzzwords such as 'tapestry', 'delve', or 'testament', to ensure that it sounds like it was written by a human. The final text must have the format <<format>> and topic <<topic>>.
67× Write a new piece of text that is on a completely different topic, but reuses verbatim as many phrases and sentences from this text as possible.
67× Write a new piece of text that uses this text as a citation or reference. The final text must have the format <<format>> and topic <<topic>>.
66× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human. The final text must have the format <<format>>.
66× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should write the new text as if it was a rough draft being typed out on a phone. The final text must have the format <<format>>.
66× Write a detailed rebuttal or critique challenging the core ideas, themes, or facts presented in this text, using the same language and roughly matching the intellectual level of the text. The final text must have the topic <<topic>>.
66× Write a piece of text in the same style and language as this text, but is directly contrasting to the themes, ideas, and/or details in this text. The final text must have the topic <<topic>>.
65× Take a subpoint within this text and write a full new text about it, using the language and wording from the text as appropriate. The final text must have the topic <<topic>>.
65× Write a piece of text in the same style and language as this text, but is directly contrasting to the themes, ideas, and/or details in this text. The final text must have the format <<format>> and topic <<topic>>.
64× Write another piece of text that resembles the style and language of this text. The final text must have the format <<format>>.
63× Write a followup to this piece of text, in the same style and language as the text.
63× Write a new piece of text that is on a completely different topic, but reuses verbatim as many phrases and sentences from this text as possible. The final text must have the format <<format>> and topic <<topic>>.
61× Take a subpoint within this text and write a full new text about it, using the language and wording from the text as appropriate. The final text must have the format <<format>>.
60× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human. The final text must have the topic <<topic>>.
60× Write a detailed rebuttal or critique challenging the core ideas, themes, or facts presented in this text, using the same language and roughly matching the intellectual level of the text. The final text must have the format <<format>> and topic <<topic>>.
59× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should write the new text as if it was a rough draft being typed out on a phone. The final text must have the format <<format>> and topic <<topic>>.
59× Write an alternate version of this text, it should follow the style and format of the original text.
58× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human.
58× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should use varied sentence lengths, and remove all common AI buzzwords such as 'tapestry', 'delve', or 'testament', to ensure that it sounds like it was written by a human. The final text must have the topic <<topic>>.
58× Write a followup to this piece of text, in the same style and language as the text. The final text must have the topic <<topic>>.
58× Write a piece of text in the same style and language as this text, but is directly contrasting to the themes, ideas, and/or details in this text.
58× Write a piece of text with a different topic that the author of this text might write about. The final text must have the topic <<topic>>.
58× Write another piece of text that is of a similar theme or background as this text.
58× Write another piece of text that resembles the style and language of this text. The final text must have the topic <<topic>>.
57× Write a new piece of text that is on a completely different topic, but reuses verbatim as many phrases and sentences from this text as possible. The final text must have the format <<format>>.
57× Write another piece of text that resembles the style and language of this text. The final text must have the format <<format>> and topic <<topic>>.
56× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should use varied sentence lengths, and remove all common AI buzzwords such as 'tapestry', 'delve', or 'testament', to ensure that it sounds like it was written by a human. The final text must have the format <<format>>.
56× Write a piece of text with a different topic that the author of this text might write about. The final text must have the format <<format>> and topic <<topic>>.
56× Write an alternate version of this text, it should follow the style and format of the original text. The final text must have the format <<format>>.
56× Write another piece of text that resembles the style and language of this text.
55× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should make sure there are slight, subtle grammatical mistakes, to make it sound like it was written by a human. The final text must have the format <<format>> and topic <<topic>>.
54× Write a new piece of text that uses this text as a citation or reference.
54× Write a new piece of text that uses this text as a citation or reference. The final text must have the format <<format>>.
54× Write a prequel to this text, describing the events, thoughts, or historical context that immediately led up to this piece being written, making sure to use similar language and wording as the author, as if it was written by the author. The final text must have the topic <<topic>>.
54× Write another piece of text that is of a similar theme or background as this text. The final text must have the topic <<topic>>.
53× Take a subpoint within this text and write a full new text about it, using the language and wording from the text as appropriate. The final text must have the format <<format>> and topic <<topic>>.
53× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should write the new text as if it was a rough draft being typed out on a phone.
53× Write a followup to this piece of text, in the same style and language as the text. The final text must have the format <<format>>.
53× Write an alternate version of this text, it should follow the style and format of the original text. The final text must have the format <<format>> and topic <<topic>>.
52× Write a piece of text with a different topic that the author of this text might write about.
51× Write a new piece of text that uses this text as a citation or reference. The final text must have the topic <<topic>>.
51× Write a piece of text in the same style and language as this text, but is directly contrasting to the themes, ideas, and/or details in this text. The final text must have the format <<format>>.
50× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should use varied sentence lengths, and remove all common AI buzzwords such as 'tapestry', 'delve', or 'testament', to ensure that it sounds like it was written by a human.
50× Write a different piece of text on the same topic. The final text must have the topic <<topic>>.
50× Write a piece of text with a different topic that the author of this text might write about. The final text must have the format <<format>>.
49× Write a prequel to this text, describing the events, thoughts, or historical context that immediately led up to this piece being written, making sure to use similar language and wording as the author, as if it was written by the author. The final text must have the format <<format>>.
48× Write a prequel to this text, describing the events, thoughts, or historical context that immediately led up to this piece being written, making sure to use similar language and wording as the author, as if it was written by the author. The final text must have the format <<format>> and topic <<topic>>.
47× Write a detailed rebuttal or critique challenging the core ideas, themes, or facts presented in this text, using the same language and roughly matching the intellectual level of the text.
47× Write an alternate version of this text, it should follow the style and format of the original text. The final text must have the topic <<topic>>.
46× Take inspiration from the style, language, and content of this text, and write a new piece of text on a topic of your choice. However, you should write the new text as if it was a rough draft being typed out on a phone. The final text must have the topic <<topic>>.
45× Write a prequel to this text, describing the events, thoughts, or historical context that immediately led up to this piece being written, making sure to use similar language and wording as the author, as if it was written by the author.
44× Write a detailed rebuttal or critique challenging the core ideas, themes, or facts presented in this text, using the same language and roughly matching the intellectual level of the text. The final text must have the format <<format>>.
42× Write a different piece of text on the same topic.
41× Take a subpoint within this text and write a full new text about it, using the language and wording from the text as appropriate.

indirect_reference3,367 rows

two-turn · 25 instructions

Show the 25 instructions ▾

201× Translate the given text to Hindi. Do not output anything besides what you were requested to write, and do not output any extra commentary.
183× Translate the given text to Spanish. Do not output anything besides what you were requested to write, and do not output any extra commentary.
181× Translate the given text to French. Do not output anything besides what you were requested to write, and do not output any extra commentary.
173× Translate the given text to Chinese. Do not output anything besides what you were requested to write, and do not output any extra commentary.
172× Translate the given text to German. Do not output anything besides what you were requested to write, and do not output any extra commentary.
154× This piece of text was cleverely generated by an LLM with a human supervising it so that the text seems authentic. Reverse engineer the process and come up with a prompt that is most probable to be the one used to generate the text. Do not output anything besides what you were requested to write, and do not output any extra commentary.
147× Formulate a set of strict, absolute constraints regarding format, word count, vocabulary restrictions, and syntax rules that, if followed perfectly by an AI, would result in generating a text nearly identical to this one. Note that your response should be formatted as a prompt that would be sent to an AI requesting it to generate something, not just a list of rules. Do not output anything besides what you were requested to write, and do not output any extra commentary.
146× Summarize this text, attempting to preserve as much of the original language of the text as possible. Do not output anything besides what you were requested to write, and do not output any extra commentary.
143× Take this text and replace one word in every four with a single underscore (use one underscore for each removed word, never several in a row). This way, it will be appropriate for a FIM task. You may decide which words to replace. Reproduce the rest of the document in full, in its original order and paragraph structure. Do not output anything besides what you were requested to write, and do not output any extra commentary.
137× Write a 'few-shot' prompt using minimal, fragmented excerpts from this text as examples. The prompt should be designed to teach an LLM the pattern needed to generate the rest of the text from scratch. Do not output anything besides what you were requested to write, and do not output any extra commentary.
134× Create the shortest possible prompt that can be sent to an LLM to reproduce this text. The prompt may include verbatim pieces of the original text, however, it must be short and accurate, working as an effective compression of the original text. Do not output anything besides what you were requested to write, and do not output any extra commentary.
133× Create a prompt that might cause an LLM to generate an output resembling this text. Do not output anything besides what you were requested to write, and do not output any extra commentary.
133× Take this text, but extract only the most meaningful sentences out of it to create a new text that is a series of disjoint sentences representing the original text. Do not output anything besides what you were requested to write, and do not output any extra commentary.
132× Replace every verb in this text with the tag [VERB], keeping all nouns, adjectives, articles and punctuation exactly as they appear. Reproduce the document in full, in its original order and paragraph structure. Do not output a legend or a list of the tags you used. Do not output anything besides what you were requested to write, and do not output any extra commentary.
128× Convert this text into a series of logical propositions or syllogisms that represent the core argument or narrative progression, stripped of all rhetorical flair. Do not output anything besides what you were requested to write, and do not output any extra commentary.
128× Create a prompt that would allow an LLM to generate this text as accurately/verbatim as possible, without the prompt containing any of the actual text. That is, you must use descriptors to accurately describe the text to generate. Do not output anything besides what you were requested to write, and do not output any extra commentary.
127× Provide only the first and last word of every sentence in this text, replacing the middle of each sentence with an ellipsis '...'. Maintain the original paragraph structure. Do not output anything besides what you were requested to write, and do not output any extra commentary.
116× Write a detailed descriptor for the style of writing of this text, including its qualitative and quantitative properties. Do not output anything besides what you were requested to write, and do not output any extra commentary.
107× Reformat this text into a sensible, structured JSON object. Do not output anything besides what you were requested to write, and do not output any extra commentary.
107× Write a detailed descriptor for how this text is stylistically differentiated from other texts, including grammatical choices and word choices not present in other texts of this kind. Do not output anything besides what you were requested to write, and do not output any extra commentary.
106× Envision the scenario in which the author wrote this text. Describe that scenario in exhaustive detail. Do not output anything besides what you were requested to write, and do not output any extra commentary.
106× Write a detailed descriptor for the manner in which the author of this uses language, and things such as their word choice. Do not output anything besides what you were requested to write, and do not output any extra commentary.
101× Write a detailed descriptor for the (imaginary) personality of the author who wrote this text, including their livelihood and day to day activities, as well as how that influences their writing. Do not output anything besides what you were requested to write, and do not output any extra commentary.
92× Create a detailed dictionary that fully describes all the meaningful phrases and words this author utilizes to create this text, and how the author utilizes those words and phrases, and why. Do not output anything besides what you were requested to write, and do not output any extra commentary.
80× Translate the entirety of this text into a sequence of emojis that captures the literal meaning, tone, and narrative progression of the writing. Do not use any letters or words. Do not output anything besides what you were requested to write, and do not output any extra commentary.

revise3,444 rows

1,727 single · 1,717 two-turn · 172 instructions

Show the 172 instructions ▾

68× Adapt this text for a teenage audience.
63× Adapt this text for a 5th-grade reading level.
63× Adapt this text for social media sharing.
59× Change the tone of this text to be more urgent and compelling.
59× Clarify this text by making the argument and main points stronger and easier to understand.
58× Change the tone of this text to be more enthusiastic and energetic.
57× Adapt this text for a lay audience with no technical background.
57× Change the tone of this text to be more relaxed and friendly.
57× Clarify this text by paraphrasing it with different vocabulary and varied sentence structures.
56× Change the tone of this text to be more inspiring and motivational.
55× Change the tone of this text to be more objective and unbiased.
55× Clarify this text by defining any terms that might be unclear to the reader.
54× Change the tone of this text to be more analytical and logical.
54× Expand this text by adding vivid imagery and sensory details to bring it to life.
53× Change the tone of this text to be more conversational and approachable.
53× Expand this text by adding supporting information to elaborate on the key points.
52× Restructure this text to read more naturally, as if written by a native English speaker.
50× Change the tone of this text to be more academic and scholarly.
50× Clarify this text by improving the phrasing while preserving its original intent.
50× Expand this text by adding backstory and context to enrich understanding.
50× Expand this text by adding storytelling elements and dialogue to increase engagement.
50× Restructure this text to arrange the points in order of importance.
48× Adapt this text for an elderly adult with limited knowledge of modern day affairs.
48× Change the tone of this text to be more empathetic and understanding.
47× Change the tone of this text to be more lighthearted and humorous.
47× Clarify this text by presenting the same information from a fresh angle.
47× Restructure this text to improve the logical flow and argument progression.
46× Restructure this text to improve the transitions and connections between ideas.
45× Change the tone of this text to be more polite and diplomatic.
45× Edit this text to eliminate redundant and filler words.
45× Expand this text by adding concrete examples to illustrate the main points.
45× Restructure this text to eliminate awkward phrasing and improve the overall rhythm.
44× Change the tone of this text to be more professional and authoritative.
44× Clarify this text by making the cause-and-effect relationships more obvious.
44× Edit this text to trim the length without losing important information or tone.
44× Restructure this text to ensure a clear introduction, body, and compelling conclusion.
43× Change the tone of this text to be more direct and bold.
43× Change the tone of this text to be more persuasive and convincing.
43× Edit this text to be more concise and powerful.
43× Expand this text by adding real-world applications and specific details.
40× Clarify this text by eliminating any ambiguous or vague language.
40× Restructure this text to group related ideas and include clear section breaks.
37× Change the tone of this text to be more suited for a business audience.
20× Expand this text by adding vivid imagery and sensory details to bring it to life. The final text must have the format <<format>>.
17× Adapt this text for a lay audience with no technical background. The final text must have the format <<format>>.
17× Change the tone of this text to be more professional and authoritative. The final text must have the format <<format>> and topic <<topic>>.
17× Expand this text by adding storytelling elements and dialogue to increase engagement. The final text must have the topic <<topic>>.
15× Change the tone of this text to be more polite and diplomatic. The final text must have the format <<format>>.
15× Edit this text to be more concise and powerful. The final text must have the format <<format>> and topic <<topic>>.
15× Expand this text by adding concrete examples to illustrate the main points. The final text must have the format <<format>> and topic <<topic>>.
15× Restructure this text to group related ideas and include clear section breaks. The final text must have the format <<format>>.
14× Adapt this text for a teenage audience. The final text must have the format <<format>>.
14× Change the tone of this text to be more analytical and logical. The final text must have the format <<format>> and topic <<topic>>.
14× Change the tone of this text to be more direct and bold. The final text must have the topic <<topic>>.
14× Change the tone of this text to be more lighthearted and humorous. The final text must have the format <<format>> and topic <<topic>>.
14× Change the tone of this text to be more urgent and compelling. The final text must have the format <<format>>.
14× Clarify this text by defining any terms that might be unclear to the reader. The final text must have the format <<format>> and topic <<topic>>.
14× Clarify this text by eliminating any ambiguous or vague language. The final text must have the topic <<topic>>.
14× Clarify this text by presenting the same information from a fresh angle. The final text must have the format <<format>> and topic <<topic>>.
14× Expand this text by adding concrete examples to illustrate the main points. The final text must have the format <<format>>.
13× Adapt this text for a 5th-grade reading level. The final text must have the format <<format>> and topic <<topic>>.
13× Change the tone of this text to be more academic and scholarly. The final text must have the topic <<topic>>.
13× Change the tone of this text to be more direct and bold. The final text must have the format <<format>>.
13× Change the tone of this text to be more empathetic and understanding. The final text must have the format <<format>>.
13× Change the tone of this text to be more empathetic and understanding. The final text must have the topic <<topic>>.
13× Change the tone of this text to be more lighthearted and humorous. The final text must have the topic <<topic>>.
13× Change the tone of this text to be more relaxed and friendly. The final text must have the topic <<topic>>.
13× Restructure this text to arrange the points in order of importance. The final text must have the format <<format>>.
13× Restructure this text to eliminate awkward phrasing and improve the overall rhythm. The final text must have the format <<format>> and topic <<topic>>.
12× Adapt this text for a 5th-grade reading level. The final text must have the format <<format>>.
12× Change the tone of this text to be more direct and bold. The final text must have the format <<format>> and topic <<topic>>.
12× Change the tone of this text to be more enthusiastic and energetic. The final text must have the topic <<topic>>.
12× Change the tone of this text to be more inspiring and motivational. The final text must have the format <<format>>.
12× Change the tone of this text to be more polite and diplomatic. The final text must have the topic <<topic>>.
12× Change the tone of this text to be more relaxed and friendly. The final text must have the format <<format>> and topic <<topic>>.
12× Change the tone of this text to be more relaxed and friendly. The final text must have the format <<format>>.
12× Change the tone of this text to be more urgent and compelling. The final text must have the format <<format>> and topic <<topic>>.
12× Clarify this text by defining any terms that might be unclear to the reader. The final text must have the format <<format>>.
12× Clarify this text by eliminating any ambiguous or vague language. The final text must have the format <<format>> and topic <<topic>>.
12× Clarify this text by making the cause-and-effect relationships more obvious. The final text must have the format <<format>> and topic <<topic>>.
12× Expand this text by adding storytelling elements and dialogue to increase engagement. The final text must have the format <<format>> and topic <<topic>>.
12× Expand this text by adding vivid imagery and sensory details to bring it to life. The final text must have the topic <<topic>>.
12× Restructure this text to ensure a clear introduction, body, and compelling conclusion. The final text must have the format <<format>> and topic <<topic>>.
12× Restructure this text to improve the transitions and connections between ideas. The final text must have the format <<format>>.
12× Restructure this text to read more naturally, as if written by a native English speaker. The final text must have the format <<format>>.
12× Restructure this text to read more naturally, as if written by a native English speaker. The final text must have the topic <<topic>>.
11× Adapt this text for a teenage audience. The final text must have the topic <<topic>>.
11× Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the topic <<topic>>.
11× Change the tone of this text to be more analytical and logical. The final text must have the format <<format>>.
11× Change the tone of this text to be more professional and authoritative. The final text must have the topic <<topic>>.
11× Change the tone of this text to be more suited for a business audience. The final text must have the format <<format>> and topic <<topic>>.
11× Clarify this text by eliminating any ambiguous or vague language. The final text must have the format <<format>>.
11× Expand this text by adding backstory and context to enrich understanding. The final text must have the format <<format>> and topic <<topic>>.
11× Expand this text by adding concrete examples to illustrate the main points. The final text must have the topic <<topic>>.
11× Restructure this text to eliminate awkward phrasing and improve the overall rhythm. The final text must have the topic <<topic>>.
11× Restructure this text to ensure a clear introduction, body, and compelling conclusion. The final text must have the format <<format>>.
11× Restructure this text to improve the transitions and connections between ideas. The final text must have the topic <<topic>>.
10× Adapt this text for a 5th-grade reading level. The final text must have the topic <<topic>>.
10× Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>> and topic <<topic>>.
10× Change the tone of this text to be more academic and scholarly. The final text must have the format <<format>>.
10× Change the tone of this text to be more enthusiastic and energetic. The final text must have the format <<format>> and topic <<topic>>.
10× Change the tone of this text to be more inspiring and motivational. The final text must have the topic <<topic>>.
10× Change the tone of this text to be more professional and authoritative. The final text must have the format <<format>>.
10× Clarify this text by making the argument and main points stronger and easier to understand. The final text must have the format <<format>> and topic <<topic>>.
10× Clarify this text by making the cause-and-effect relationships more obvious. The final text must have the format <<format>>.
10× Clarify this text by making the cause-and-effect relationships more obvious. The final text must have the topic <<topic>>.
10× Edit this text to be more concise and powerful. The final text must have the format <<format>>.
10× Edit this text to eliminate redundant and filler words. The final text must have the format <<format>> and topic <<topic>>.
10× Edit this text to eliminate redundant and filler words. The final text must have the topic <<topic>>.
10× Expand this text by adding real-world applications and specific details. The final text must have the topic <<topic>>.
10× Expand this text by adding supporting information to elaborate on the key points. The final text must have the format <<format>> and topic <<topic>>.
10× Expand this text by adding vivid imagery and sensory details to bring it to life. The final text must have the format <<format>> and topic <<topic>>.
10× Restructure this text to improve the logical flow and argument progression. The final text must have the format <<format>> and topic <<topic>>.
10× Restructure this text to improve the logical flow and argument progression. The final text must have the format <<format>>.
10× Restructure this text to improve the logical flow and argument progression. The final text must have the topic <<topic>>.
10× Restructure this text to improve the transitions and connections between ideas. The final text must have the format <<format>> and topic <<topic>>.
9× Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>>.
9× Adapt this text for social media sharing. The final text must have the format <<format>> and topic <<topic>>.
9× Change the tone of this text to be more conversational and approachable. The final text must have the format <<format>>.
9× Change the tone of this text to be more conversational and approachable. The final text must have the topic <<topic>>.
9× Change the tone of this text to be more empathetic and understanding. The final text must have the format <<format>> and topic <<topic>>.
9× Change the tone of this text to be more enthusiastic and energetic. The final text must have the format <<format>>.
9× Change the tone of this text to be more lighthearted and humorous. The final text must have the format <<format>>.
9× Change the tone of this text to be more objective and unbiased. The final text must have the topic <<topic>>.
9× Change the tone of this text to be more urgent and compelling. The final text must have the topic <<topic>>.
9× Clarify this text by defining any terms that might be unclear to the reader. The final text must have the topic <<topic>>.
9× Clarify this text by improving the phrasing while preserving its original intent. The final text must have the topic <<topic>>.
9× Clarify this text by making the argument and main points stronger and easier to understand. The final text must have the format <<format>>.
9× Clarify this text by paraphrasing it with different vocabulary and varied sentence structures. The final text must have the format <<format>>.
9× Clarify this text by paraphrasing it with different vocabulary and varied sentence structures. The final text must have the topic <<topic>>.
9× Clarify this text by presenting the same information from a fresh angle. The final text must have the topic <<topic>>.
9× Edit this text to trim the length without losing important information or tone. The final text must have the topic <<topic>>.
9× Restructure this text to group related ideas and include clear section breaks. The final text must have the format <<format>> and topic <<topic>>.
8× Adapt this text for a lay audience with no technical background. The final text must have the format <<format>> and topic <<topic>>.
8× Adapt this text for a teenage audience. The final text must have the format <<format>> and topic <<topic>>.
8× Adapt this text for an elderly adult with limited knowledge of modern day affairs. The final text must have the format <<format>> and topic <<topic>>.
8× Change the tone of this text to be more conversational and approachable. The final text must have the format <<format>> and topic <<topic>>.
8× Change the tone of this text to be more inspiring and motivational. The final text must have the format <<format>> and topic <<topic>>.
8× Change the tone of this text to be more objective and unbiased. The final text must have the format <<format>> and topic <<topic>>.
8× Change the tone of this text to be more persuasive and convincing. The final text must have the topic <<topic>>.
8× Change the tone of this text to be more suited for a business audience. The final text must have the topic <<topic>>.
8× Clarify this text by making the argument and main points stronger and easier to understand. The final text must have the topic <<topic>>.
8× Expand this text by adding backstory and context to enrich understanding. The final text must have the format <<format>>.
8× Expand this text by adding real-world applications and specific details. The final text must have the format <<format>> and topic <<topic>>.
8× Expand this text by adding storytelling elements and dialogue to increase engagement. The final text must have the format <<format>>.
8× Restructure this text to eliminate awkward phrasing and improve the overall rhythm. The final text must have the format <<format>>.
8× Restructure this text to ensure a clear introduction, body, and compelling conclusion. The final text must have the topic <<topic>>.
8× Restructure this text to read more naturally, as if written by a native English speaker. The final text must have the format <<format>> and topic <<topic>>.
7× Adapt this text for social media sharing. The final text must have the topic <<topic>>.
7× Change the tone of this text to be more analytical and logical. The final text must have the topic <<topic>>.
7× Change the tone of this text to be more persuasive and convincing. The final text must have the format <<format>> and topic <<topic>>.
7× Change the tone of this text to be more persuasive and convincing. The final text must have the format <<format>>.
7× Change the tone of this text to be more polite and diplomatic. The final text must have the format <<format>> and topic <<topic>>.
7× Change the tone of this text to be more suited for a business audience. The final text must have the format <<format>>.
7× Edit this text to eliminate redundant and filler words. The final text must have the format <<format>>.
6× Adapt this text for social media sharing. The final text must have the format <<format>>.
6× Change the tone of this text to be more objective and unbiased. The final text must have the format <<format>>.
6× Clarify this text by paraphrasing it with different vocabulary and varied sentence structures. The final text must have the format <<format>> and topic <<topic>>.
6× Clarify this text by presenting the same information from a fresh angle. The final text must have the format <<format>>.
6× Edit this text to trim the length without losing important information or tone. The final text must have the format <<format>> and topic <<topic>>.
6× Expand this text by adding supporting information to elaborate on the key points. The final text must have the topic <<topic>>.
6× Restructure this text to group related ideas and include clear section breaks. The final text must have the topic <<topic>>.
5× Clarify this text by improving the phrasing while preserving its original intent. The final text must have the format <<format>>.
5× Expand this text by adding supporting information to elaborate on the key points. The final text must have the format <<format>>.
5× Restructure this text to arrange the points in order of importance. The final text must have the topic <<topic>>.
4× Adapt this text for a lay audience with no technical background. The final text must have the topic <<topic>>.
4× Clarify this text by improving the phrasing while preserving its original intent. The final text must have the format <<format>> and topic <<topic>>.
4× Edit this text to trim the length without losing important information or tone. The final text must have the format <<format>>.
4× Expand this text by adding backstory and context to enrich understanding. The final text must have the topic <<topic>>.
4× Restructure this text to arrange the points in order of importance. The final text must have the format <<format>> and topic <<topic>>.
3× Expand this text by adding real-world applications and specific details. The final text must have the format <<format>>.
2× Edit this text to be more concise and powerful. The final text must have the topic <<topic>>.

rewrite3,290 rows

1,622 single · 1,668 two-turn · 104 instructions

Show the 104 instructions ▾

94× Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else.
91× Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow.
90× Edit this text by inserting twenty to thirty words (in appropriate locations), without changing anything else.
88× Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow.
87× Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows.
87× Edit this text by inserting a singular new section (in an appropriate locations), you may only make minimal modifications to transitions in adjacent sections for better flow.
87× Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text.
86× Modify this text so that it is as short as possible. At least a quarter of the original text must be repeated verbatim, and at most half of it; the remainder must be new prose.
83× Rewrite the weakest section in this text, but do not make any modifications to any other section.
82× Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched.
82× Edit this text by replacing all instances of passive voice with active voice, without altering any sentences that are already in the active voice.
82× Improve this text by rewriting it. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose.
81× Modify this text to be as long as possible. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose.
80× Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow.
80× Edit this text by replacing exactly one to three sentences, without changing anything else.
78× Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound.
78× Rewrite exactly one section of this text so that its wording is entirely new, and leave every other section untouched. The rewritten section must share no sentence with the original.
76× Edit this text by replacing, removing or inserting twenty to thirty words, without changing anything else.
75× Edit this text by completely rewriting the topic sentence of every paragraph, while preserving the supporting sentences in those paragraphs verbatim.
74× Edit this text by replacing generic adjectives with highly specific, evocative ones, modifying no more than ten percent of the total words.
74× Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text.
72× Edit this text by replacing five to ten percent of words in it, without changing anything else.
67× Rewrite only the opening hook and the final concluding sentence of this text, leaving the entire middle portion exactly as it was originally written.
66× Edit this text by replacing all existing transition words and phrases (e.g., 'however', 'therefore', 'in addition') with different, synonymous transitions, making no other changes.
64× Edit this text by upgrading the verbs to be more dynamic and action-oriented, leaving the subjects, objects, and overall length completely unchanged.
58× Edit this text by removing five to ten percent of the words from it (the least necessary), without changing anything else.
27× Modify this text to be as long as possible. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the format <<format>> and topic <<topic>>.
26× Improve this text by rewriting it. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the topic <<topic>>.
25× Edit this text by removing five to ten percent of the words from it (the least necessary), without changing anything else. The final text must have the format <<format>>.
24× Edit this text by replacing generic adjectives with highly specific, evocative ones, modifying no more than ten percent of the total words. The final text must have the format <<format>> and topic <<topic>>.
23× Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the format <<format>> and topic <<topic>>.
22× Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the format <<format>> and topic <<topic>>.
22× Edit this text by inserting twenty to thirty words (in appropriate locations), without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
22× Rewrite exactly one section of this text so that its wording is entirely new, and leave every other section untouched. The rewritten section must share no sentence with the original. The final text must have the topic <<topic>>.
21× Edit this text by upgrading the verbs to be more dynamic and action-oriented, leaving the subjects, objects, and overall length completely unchanged. The final text must have the format <<format>>.
20× Edit this text by replacing all existing transition words and phrases (e.g., 'however', 'therefore', 'in addition') with different, synonymous transitions, making no other changes. The final text must have the format <<format>> and topic <<topic>>.
20× Edit this text by replacing all instances of passive voice with active voice, without altering any sentences that are already in the active voice. The final text must have the format <<format>>.
20× Edit this text by replacing, removing or inserting twenty to thirty words, without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
20× Modify this text so that it is as short as possible. At least a quarter of the original text must be repeated verbatim, and at most half of it; the remainder must be new prose. The final text must have the format <<format>> and topic <<topic>>.
19× Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched. The final text must have the topic <<topic>>.
19× Edit this text by completely rewriting the topic sentence of every paragraph, while preserving the supporting sentences in those paragraphs verbatim. The final text must have the format <<format>> and topic <<topic>>.
19× Edit this text by completely rewriting the topic sentence of every paragraph, while preserving the supporting sentences in those paragraphs verbatim. The final text must have the format <<format>>.
19× Edit this text by removing five to ten percent of the words from it (the least necessary), without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
19× Edit this text by replacing five to ten percent of words in it, without changing anything else. The final text must have the format <<format>>.
19× Edit this text by replacing, removing or inserting twenty to thirty words, without changing anything else. The final text must have the format <<format>>.
18× Edit this text by inserting twenty to thirty words (in appropriate locations), without changing anything else. The final text must have the topic <<topic>>.
18× Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the topic <<topic>>.
18× Modify this text so that it is as short as possible. At least a quarter of the original text must be repeated verbatim, and at most half of it; the remainder must be new prose. The final text must have the format <<format>>.
18× Rewrite exactly one section of this text so that its wording is entirely new, and leave every other section untouched. The rewritten section must share no sentence with the original. The final text must have the format <<format>>.
18× Rewrite the weakest section in this text, but do not make any modifications to any other section. The final text must have the topic <<topic>>.
17× Edit this text by replacing all instances of passive voice with active voice, without altering any sentences that are already in the active voice. The final text must have the topic <<topic>>.
17× Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the format <<format>> and topic <<topic>>.
17× Modify this text to be as long as possible. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the format <<format>>.
17× Rewrite exactly one section of this text so that its wording is entirely new, and leave every other section untouched. The rewritten section must share no sentence with the original. The final text must have the format <<format>> and topic <<topic>>.
16× Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>>.
16× Edit this text by inserting a singular new section (in an appropriate locations), you may only make minimal modifications to transitions in adjacent sections for better flow. The final text must have the format <<format>>.
16× Edit this text by replacing exactly one to three sentences, without changing anything else. The final text must have the topic <<topic>>.
16× Edit this text by upgrading the verbs to be more dynamic and action-oriented, leaving the subjects, objects, and overall length completely unchanged. The final text must have the topic <<topic>>.
16× Improve this text by rewriting it. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the format <<format>>.
15× Edit this text by completely rewriting the topic sentence of every paragraph, while preserving the supporting sentences in those paragraphs verbatim. The final text must have the topic <<topic>>.
15× Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the format <<format>>.
15× Edit this text by removing five to ten percent of the words from it (the least necessary), without changing anything else. The final text must have the topic <<topic>>.
15× Edit this text by replacing all existing transition words and phrases (e.g., 'however', 'therefore', 'in addition') with different, synonymous transitions, making no other changes. The final text must have the format <<format>>.
15× Edit this text by replacing the final sentence with a more open-ended, thought-provoking statement, without altering any preceding text. The final text must have the format <<format>>.
15× Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
15× Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else. The final text must have the topic <<topic>>.
15× Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>> and topic <<topic>>.
15× Edit this text by upgrading the verbs to be more dynamic and action-oriented, leaving the subjects, objects, and overall length completely unchanged. The final text must have the format <<format>> and topic <<topic>>.
15× Rewrite only the opening hook and the final concluding sentence of this text, leaving the entire middle portion exactly as it was originally written. The final text must have the format <<format>> and topic <<topic>>.
15× Rewrite only the opening hook and the final concluding sentence of this text, leaving the entire middle portion exactly as it was originally written. The final text must have the format <<format>>.
15× Rewrite only the opening hook and the final concluding sentence of this text, leaving the entire middle portion exactly as it was originally written. The final text must have the topic <<topic>>.
15× Rewrite the weakest section in this text, but do not make any modifications to any other section. The final text must have the format <<format>> and topic <<topic>>.
15× Rewrite the weakest section in this text, but do not make any modifications to any other section. The final text must have the format <<format>>.
14× Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched. The final text must have the format <<format>> and topic <<topic>>.
14× Edit this text by inserting a singular new section (in an appropriate locations), you may only make minimal modifications to transitions in adjacent sections for better flow. The final text must have the format <<format>> and topic <<topic>>.
14× Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the format <<format>>.
14× Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow. The final text must have the format <<format>> and topic <<topic>>.
14× Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow. The final text must have the format <<format>>.
14× Edit this text by removing one section from it (the least necessary), you may only make minimal modifications to adjacent sections for better flow. The final text must have the topic <<topic>>.
14× Edit this text by replacing all existing transition words and phrases (e.g., 'however', 'therefore', 'in addition') with different, synonymous transitions, making no other changes. The final text must have the topic <<topic>>.
14× Edit this text by replacing exactly one to three sentences, without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
14× Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the format <<format>>.
13× Edit this text by combining at least four pairs of short, adjacent sentences into complex sentences using appropriate conjunctions, leaving all other sentences untouched. The final text must have the format <<format>>.
13× Edit this text by inserting a singular new section (in an appropriate locations), you may only make minimal modifications to transitions in adjacent sections for better flow. The final text must have the topic <<topic>>.
13× Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the topic <<topic>>.
13× Edit this text by removing exactly one to three sentences from it (the least necessary), you may only make minimal modifications to adjacent sentences for better flow. The final text must have the topic <<topic>>.
13× Edit this text by replacing generic adjectives with highly specific, evocative ones, modifying no more than ten percent of the total words. The final text must have the topic <<topic>>.
13× Edit this text by replacing, removing or inserting exactly one to three sentences, without changing anything else. The final text must have the format <<format>>.
13× Edit this text by splitting the three longest sentences into shorter, distinct sentences, without changing the vocabulary or structure of the rest of the text. The final text must have the topic <<topic>>.
12× Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the format <<format>> and topic <<topic>>.
12× Edit this text by inserting exactly one to three sentences (in appropriate locations), you may only make minimal modifications to transitions in adjacent sentences for better flow. The final text must have the topic <<topic>>.
12× Edit this text by replacing exactly one to three sentences, without changing anything else. The final text must have the format <<format>>.
12× Edit this text by replacing five to ten percent of words in it, without changing anything else. The final text must have the topic <<topic>>.
12× Improve this text by rewriting it. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the format <<format>> and topic <<topic>>.
12× Modify this text so that it is as short as possible. At least a quarter of the original text must be repeated verbatim, and at most half of it; the remainder must be new prose. The final text must have the topic <<topic>>.
11× Edit this text by inserting twenty to thirty words (in appropriate locations), without changing anything else. The final text must have the format <<format>>.
11× Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the format <<format>>.
11× Edit this text by replacing, removing or inserting twenty to thirty words, without changing anything else. The final text must have the topic <<topic>>.
10× Edit this text by replacing generic adjectives with highly specific, evocative ones, modifying no more than ten percent of the total words. The final text must have the format <<format>>.
9× Edit this text by removing all adverbs, making only minimal adjustments to the surrounding grammar to ensure the sentences remain structurally sound. The final text must have the format <<format>> and topic <<topic>>.
9× Edit this text by replacing all instances of passive voice with active voice, without altering any sentences that are already in the active voice. The final text must have the format <<format>> and topic <<topic>>.
9× Edit this text by replacing five to ten percent of words in it, without changing anything else. The final text must have the format <<format>> and topic <<topic>>.
9× Modify this text to be as long as possible. At least half of the original text must be repeated verbatim, and at most three quarters of it; the remainder must be new prose. The final text must have the topic <<topic>>.
6× Edit this text by inserting a single, highly engaging introductory sentence at the very beginning, without modifying any of the original text that follows. The final text must have the topic <<topic>>.

Topics

Topic	Rows	Share
Industrial	673	5.0%
Transportation	664	4.9%
Travel	651	4.8%
Art & Design	628	4.6%
Fashion & Beauty	624	4.6%
Hardware	623	4.6%
Software Dev.	616	4.5%
Social Life	609	4.5%
Food & Dining	606	4.5%
Sports & Fitness	595	4.4%
Religion	591	4.4%
Games	587	4.3%
Crime & Law	586	4.3%
Home & Hobbies	585	4.3%
Entertainment	583	4.3%
Education & Jobs	570	4.2%
Software	560	4.1%
Literature	559	4.1%
Politics	556	4.1%
History	555	4.1%
Health	529	3.9%
Science & Tech.	518	3.8%
Finance & Business	487	3.6%

Formats

Format	Rows	Share
Personal Blog	959	7.1%
News (Org.)	880	6.5%
About (Org.)	854	6.3%
Comment Section	852	6.3%
News Article	830	6.1%
Product Page	830	6.1%
About (Pers.)	815	6.0%
Tutorial	793	5.9%
Listicle	792	5.8%
Creative Writing	773	5.7%
Knowledge Article	767	5.7%
User Review	767	5.7%
Nonfiction Writing	758	5.6%
Academic Writing	625	4.6%
Audio Transcript	610	4.5%
Customer Support	595	4.4%
Q&A Forum	461	3.4%
FAQ	321	2.4%
Legal Notices	199	1.5%
Documentation	64	0.5%
Spam / Ads	10	0.1%

Evaluation protocol

Classes

original is Human, final_response is AI. Each detector scores both texts of every row.

Thresholds

All 13,555 rows are used; there is no sweep/evaluation split. Each detector's thresholds are placed exactly on its human scores so that 1% and 0.1% of human texts are called AI. 9 detectors call lower scores AI (for example, perplexity).

Metrics

N (texts scored), AUROC, and TPR at the 1% and 0.1% FPR thresholds. Subsets are scored at the corpus-wide thresholds rather than refitted.

Filters

None. No cosine, soft-ngram, Jaccard, or other analysis-time filtering is applied.

Skipped

None

Config

config/analysis.toml

Schema & usage57 columns · loading snippet⌄

57 columns per row. Detector columns come in pairs, one per text of the row.

Text

original

Human-written source (class: Human)

final_response

Generated text (class: AI)

response_0

Generator output for the first turn

response_1

Generator output for the second turn (empty for single-turn prompts)

Generation

prompt

struct: chat_turns (list of strings), use_multiturn, examples, metadata.PROMPT_TYPE

generator_model

Hugging Face id of the generator model

generation_params

JSON string of sampling parameters

Metadata

topic

Topic of the human source

format

Format of the human source

Distances

jaccard_1, jaccard_2, levenshtein, softngram, cosdist, bertscore, bertscore_precision, bertscore_recall, moverscore, reranker

10 float columns measuring how far the rewrite moved from the original

Other columns

original_editlens_bucket_roberta_large, final_response_editlens_bucket_roberta_large, original_editlens_bucket_llama_3_2_3b, final_response_editlens_bucket_llama_3_2_3b, original_editlens_bucket_giga_roberta, final_response_editlens_bucket_giga_roberta, original_editlens_bucket_giga_llama_3_2_3b, final_response_editlens_bucket_giga_llama_3_2_3b

Not used by this report

Detector scores

One float column per text: {original|final_response}<suffix>

_editlens_score_roberta_large

EditLens Roberta-Large Score

_editlens_score_llama_3_2_3b

EditLens Llama-3.2-3B Score

_editlens_score_giga_roberta

Giga EditLens Roberta Score

_editlens_score_giga_llama_3_2_3b

Giga EditLens Llama-3.2-3B Score

_perplexity_llama_instruct

Perplexity (Llama-3.2-3B-Instruct)

_perplexity_llama_base

Perplexity (Llama-3.2-3B)

_entropy_llama_instruct

Entropy (Llama-3.2-3B-Instruct)

_entropy_llama_base

Entropy (Llama-3.2-3B)

_topp_outlier_llama_instruct

Top-p Outliers (Llama-3.2-3B-Instruct)

_topp_outlier_llama_base

Top-p Outliers (Llama-3.2-3B)

_topk_outlier_llama_instruct

Top-k Outliers (Llama-3.2-3B-Instruct)

_topk_outlier_llama_base

Top-k Outliers (Llama-3.2-3B)

_fastdetectgpt_llama_instruct

FastDetectGPT (Llama-3.2-3B-Instruct)

_fastdetectgpt_llama_base

FastDetectGPT (Llama-3.2-3B)

_binoculars

Binoculars

Load it

from datasets import concatenate_datasets, get_dataset_config_names, load_dataset

repo = "G-reen/fastdetector-val-stat-val"
# every generator config is its own dataset config
ds = concatenate_datasets([
    load_dataset(repo, config, split="train")
    for config in get_dataset_config_names(repo)
])

Univariate statistics40 statistics · all 13,555 rows valid⌄

Every statistic the report does arithmetic on, over the complete corpus. Missing or non-finite values are counted as invalid.

Statistic	Mean	Median	Std	Min	Max
`jaccard_1`	0.6244	0.7341	0.2700	0.0000	1.0000
`jaccard_2`	0.7415	0.8881	0.2884	0.0195	1.0000
`levenshtein`	1820.0764	1206.0000	2269.6051	1.0000	65876.0000
`softngram`	0.5569	0.6184	0.3444	0.0000	1.0000
`cosdist`	0.1989	0.1284	0.2012	-0.0061	1.0315
`bertscore`	0.1338	0.1425	0.0754	0.0009	0.4338
`bertscore_precision`	0.1315	0.1384	0.0746	0.0007	0.5106
`bertscore_recall`	0.1350	0.1416	0.0813	0.0004	0.4319
`moverscore`	0.5252	0.5801	0.1929	0.0559	1.0810
`reranker`	-3.3201	-5.3750	5.7513	-11.1875	19.0000
`EditLens Roberta-Large Score (Human)`	0.0580	0.0205	0.0981	0.0064	0.9993
`EditLens Roberta-Large Score (AI)`	0.4699	0.3776	0.3735	0.0064	0.9996
`EditLens Llama-3.2-3B Score (Human)`	0.0456	0.0226	0.0623	0.0003	0.8558
`EditLens Llama-3.2-3B Score (AI)`	0.4867	0.4493	0.3523	0.0007	1.0000
`Giga EditLens Roberta Score (Human)`	0.0255	0.0048	0.0957	0.0013	0.9999
`Giga EditLens Roberta Score (AI)`	0.5424	0.5882	0.4130	0.0013	1.0000
`Giga EditLens Llama-3.2-3B Score (Human)`	0.0119	0.0038	0.0368	0.0004	0.9339
`Giga EditLens Llama-3.2-3B Score (AI)`	0.4690	0.4111	0.3852	0.0005	1.0000
`Perplexity (Llama-3.2-3B-Instruct) (Human)`	19.3247	15.9592	15.6930	1.6941	757.4117
`Perplexity (Llama-3.2-3B-Instruct) (AI)`	15.5311	12.4861	17.1405	1.0586	990.4898
`Perplexity (Llama-3.2-3B) (Human)`	14.3750	12.3143	10.0975	1.0484	485.7397
`Perplexity (Llama-3.2-3B) (AI)`	12.8630	10.5607	12.1584	1.0502	551.9839
`Entropy (Llama-3.2-3B-Instruct) (Human)`	2.6211	2.5815	0.4935	0.4899	7.6310
`Entropy (Llama-3.2-3B-Instruct) (AI)`	2.3292	2.3373	0.6299	0.0945	6.7115
`Entropy (Llama-3.2-3B) (Human)`	2.5188	2.5096	0.4546	0.0784	5.6794
`Entropy (Llama-3.2-3B) (AI)`	2.3723	2.3829	0.5169	0.0607	6.0152
`Top-p Outliers (Llama-3.2-3B-Instruct) (Human)`	0.0583	0.0554	0.0170	0.0000	0.1746
`Top-p Outliers (Llama-3.2-3B-Instruct) (AI)`	0.0558	0.0537	0.0181	0.0000	0.2000
`Top-p Outliers (Llama-3.2-3B) (Human)`	0.0430	0.0426	0.0128	0.0000	0.1304
`Top-p Outliers (Llama-3.2-3B) (AI)`	0.0389	0.0385	0.0153	0.0000	0.1667
`Top-k Outliers (Llama-3.2-3B-Instruct) (Human)`	0.1071	0.1005	0.0447	0.0000	0.5462
`Top-k Outliers (Llama-3.2-3B-Instruct) (AI)`	0.0894	0.0827	0.0513	0.0000	0.5481
`Top-k Outliers (Llama-3.2-3B) (Human)`	0.0878	0.0817	0.0404	0.0000	0.5422
`Top-k Outliers (Llama-3.2-3B) (AI)`	0.0787	0.0716	0.0469	0.0000	0.5294
`FastDetectGPT (Llama-3.2-3B-Instruct) (Human)`	-1.6732	-1.6649	1.6447	-15.3432	25.5512
`FastDetectGPT (Llama-3.2-3B-Instruct) (AI)`	-1.6257	-1.6140	1.6757	-15.5276	10.1506
`FastDetectGPT (Llama-3.2-3B) (Human)`	-0.1269	-0.0940	1.0514	-6.4372	5.8209
`FastDetectGPT (Llama-3.2-3B) (AI)`	0.1968	0.1578	1.4935	-15.9713	10.6377
`Binoculars (Human)`	0.8549	0.8634	0.0989	0.0074	1.1340
`Binoculars (AI)`	0.8812	0.8816	0.0738	0.0450	1.2209

Correlation heatmappairwise Pearson r⌄

Each pair is correlated over the rows where both are finite.Open full size ↗

Generated by the FastDetector analysis pipeline from the complete published corpus.config/analysis.toml

Downloads last month: 1,555