[{"data":1,"prerenderedAt":259},["ShallowReactive",2],{"project-23075893":3},{"id":4,"name":5,"line":6,"line_source":7,"title":8,"url":9,"domain":5,"maker":10,"posted_at":11,"year":12,"points":13,"comments":14,"rank_in_year":15,"shown_in_year":16,"top":17,"tier":18,"category":19,"subcategory":20,"format":21,"keywords":22,"reception":28,"sentiment":29,"stance_mix":30,"scored_comments":34,"aspects":35,"strip":40,"post_text":42,"what":43,"said":69,"maker_in_thread":151,"shape":160,"grid":164,"attempts":222,"same_week":223,"maker_shows":215,"maker_hidden":252,"reader_version":253,"card":254},"23075893","goopt2.xyz","I trained GPT-2 to write like Goop","title","Show HN: I trained GPT-2 to write like Goop","http:\u002F\u002Fgoopt2.xyz\u002F","calebkaiser",1588647029,2020,96,25,671,3028,23,"notable","learning_science","ml text generation demo","web_app",[23,24,25,26,27],"gpt-2","text generation","goop","language model","quiz","curious",0,{"positive":31,"neutral":32,"mixed":29,"critical":33},3,7,2,12,{"execution":36,"idea":38,"docs":39},{"positive":29,"critical":37},1,{"positive":33,"critical":29},{"positive":37,"critical":29},[37,29,29,29,29,31,31,29,29,31,29,41,31,41,29,29,29,29,29,31,29,41,31,31,37],5,null,{"problem":44,"audience":42,"pricing":42,"motivation":51,"time_to_build":42,"stack":58,"openness":67,"stage":68,"team_size":67},{"value":45,"source":46,"evidence":47},"Shows how hard it is to tell GPT-2 generated text from real Goop wellness articles","maker",[48],{"id":49,"by":10,"text":50},"23075922","Hey HN -- I had a slow weekend and built this thing. It's a little game that displays a sentence, and you decide if you think an ML model generated it, or if it's an actual quote…",{"value":52,"source":46,"evidence":53},"Weekend fun project poking fun at Goop",[54,55],{"id":49,"by":10,"text":50},{"id":56,"by":10,"text":57},"23076654","Thanks for this—I probably should have clarified. Goop is a psuedoscience-as-a-lifestyle brand that frequently publishes hilariously terrible medical advice under the guise of…",[59,63],{"value":60,"source":46,"evidence":61},"GPT-2 (OpenAI)",[62],{"id":49,"by":10,"text":50},{"value":64,"source":46,"evidence":65},"gpt-2-simple (Python package)",[66],{"id":49,"by":10,"text":50},"unknown","launched",{"praise":70,"objections":90,"requests":108,"questions":123,"alternatives":136},[71,83],{"value":72,"source":73,"evidence":74},"Fun and funny; the Goop-versus-GPT-2 quiz is entertaining","commenter",[75,79],{"id":76,"by":77,"text":78},"23076464","elil17","I can't believe how bad I am with this! It reminds me of the book \"The Most Human Human.\" Every year, at the competition for the Loebner Prize (a five minute chatbot Turing…",{"id":80,"by":81,"text":82},"23078654","avianlyric","Looking at the site, and the authors comment, this looks like a quick weekend project made for fun. I wouldn’t spend too much time reading into it or critiquing it’s honesty. It…",{"value":84,"source":73,"evidence":85},"Good idea for crowdsourced labeling of data",[86],{"id":87,"by":88,"text":89},"23078093","r0b05","Brilliant work. The way you packaged it can be used to label data to improve ML models. I am thinking of such a link being sent out to numerous people to crowdsource labeling.…",[91,102],{"value":92,"source":73,"evidence":93},"Quotes look cherry-picked, so the results are biased",[94,98],{"id":95,"by":96,"text":97},"23076837","c3534l","After trying a few to get a handle on things, I decided to make guesses based solely on if the sentence makes coherent sense. I judged 14 quotes before getting a repeat, having…",{"id":99,"by":100,"text":101},"23076880","woah","Heavy curation is unfortunately rampant in the genre of \"wow check out this AI\"",{"value":103,"source":46,"evidence":104},"Server was overloaded by HN traffic",[105],{"id":106,"by":10,"text":107},"23076538","Hey everyone—looks like HN traffic squeezed my server a little more than I expected. It's upgrading now, but might not be available for about 5 minutes. Sorry about that, and…",[109,113],{"value":110,"source":46,"evidence":111},"Add more quotes to the quiz",[112],{"id":49,"by":10,"text":50},{"value":114,"source":73,"evidence":115},"Publish a blog post on how it was built",[116,120],{"id":117,"by":118,"text":119},"23080151","bulldog13","I would be interested in learning how you built it, do you have a blog entry about fine tuning the gpt-2 with the scraped text ? Or can you recommend a blog post that does…",{"id":121,"by":118,"text":122},"23081147","Yes, very much so. Good luck and thank you!",[124,132],{"value":125,"source":73,"evidence":126},"How was the GPT-2 model fine-tuned?",[127,128],{"id":117,"by":118,"text":119},{"id":129,"by":130,"text":131},"23081569","minimaxir","See the notes and Colab notebook in the source project repo: https:\u002F\u002Fgithub.com\u002Fminimaxir\u002Fgpt-2-simple I am working on a new text generation package which should be even more…",{"value":133,"source":73,"evidence":134},"Does it use common sense?",[135],{"id":87,"by":88,"text":89},[137,144],{"value":138,"source":73,"evidence":139},"SNARXiv (snarxiv.org), a parody of arXiv",[140],{"id":141,"by":142,"text":143},"23078895","martopix","Very similar: http:\u002F\u002Fsnarxiv.org\u002Fvs-arxiv\u002F",{"value":145,"source":73,"evidence":146},"MAD Magazine catalog parodies",[147],{"id":148,"by":149,"text":150},"23076940","DrScump","This reminds me of the time that MAD Magazine did parodies of entries in the old Spencer Gifts catalogs.",{"comments":41,"answered":37,"key":152},[153,154,157],{"id":49,"text":50},{"id":155,"text":156},"23080526","Thanks! I'm planning on writing something up this week. I can message you when it's done, if you're interested.",{"id":158,"text":159},"23076550","100%. Part of what made it so fun to generate the text was that all the usual data points I instinctively use to vet ML-generated text get thrown out the window with Goop. A Goop…",{"commenters":161,"max_depth":162,"maker_replies":41,"first_comment_mins":31,"lifespan_hours":163},19,4,13,[165,167,169,172,175,177,181,183,186,188,190,191,193,194,196,197,201,202,205,208,211,213,214,216,217],[95,29,37,166,96,29],"project\u002Fcritical\u002Fexecution",[99,29,29,168,100,29],"commenter\u002Fneutral\u002Fnone",[170,29,29,168,171,29],"23078307","quickthrower2",[173,29,29,168,174,29],"23079307","craftinator",[176,29,29,168,130,29],"23081514",[178,29,31,179,180,29],"23077729","project\u002Fneutral\u002Fexecution","arcturus17",[80,29,31,182,81,29],"project\u002Fneutral\u002Fidea",[184,29,29,168,185,29],"23079276","zests",[49,37,29,187,10,37],"maker\u002Fneutral\u002Fidea",[117,37,31,189,118,29],"maker\u002Fneutral\u002Fdocs",[155,37,29,189,10,37],[121,37,41,192,118,29],"maker\u002Fpositive\u002Fdocs",[129,37,31,189,130,29],[76,33,41,195,77,29],"project\u002Fpositive\u002Fidea",[158,33,29,195,10,37],[198,31,29,199,200,29],"23076603","other\u002Fneutral\u002Fnone","mattigames",[56,31,29,199,10,37],[203,31,29,199,204,29],"23076836","pests",[206,31,29,199,207,29],"23079265","tyingq",[209,162,31,182,210,29],"23078885","andy_ppp",[106,41,29,212,10,37],"maker\u002Fneutral\u002Fperformance",[87,41,41,195,88,29],[141,215,31,182,142,29],6,[148,32,31,182,149,29],[218,219,37,220,221,29],"23077016",8,"project\u002Fcritical\u002Fnone","fake-name",[],[224,229,234,238,242,247],{"id":225,"name":226,"points":227,"category":228},"23045207","uPlot",785,"data",{"id":230,"name":231,"points":232,"category":233},"23070567","Caddy 2",778,"infrastructure",{"id":235,"name":236,"points":237,"category":19},"23066776","SQL Police Department",550,{"id":239,"name":240,"points":241,"category":19},"23042079","RegEx for Regular Folk",502,{"id":243,"name":244,"points":245,"category":246},"23033387","Tara",406,"productivity",{"id":248,"name":249,"points":250,"category":251},"23042558","Ameelio.org",364,"web_saas",false,"v2",{"id":255,"name":5,"line":6,"ai":252,"cat":19,"year":12,"points":13,"comments":14,"rec":28,"s":29,"top":17,"kw":256,"maker":10,"domain":5,"tier":18,"attempts":29,"strip":257,"url":9,"mix":258,"praise":72,"objection":92,"sub":20},23075893,[23,24,25,26,27],"1000033003053500000305331",[31,32,29,33],1794036590037]