{"id":"5338e4aeb4b9","page":"/c/5338e4aeb4b9/","topic":"Internal OpenAI Instructions following eval accuracy (hard subset)","publisher":"OpenAI","official":true,"platform":"X","publishedAt":"2025-04-14T19:12:23.000Z","model":"GPT-4","form":"benchmark-bars","metrics":{"views":16643,"likes":84,"reposts":6,"quotes":1,"replies":1,"saves":9},"readAt":"2026-10-12T21:30:08.589Z","chartsInPost":1,"score":35,"scoreWeights":{"views":0.4,"likes":0.3,"eng":0.15,"save":0.15},"quadrant":"quiet","ranks":{"score":{"industry":{"n":819,"of":1162},"company":{"n":71,"of":74}},"views":{"industry":{"n":740,"of":1162},"company":{"n":70,"of":74}},"likes":{"industry":{"n":783,"of":1162},"company":{"n":72,"of":74}},"eng":{"industry":{"n":767,"of":1162},"company":{"n":36,"of":74}},"save":{"industry":{"n":770,"of":1162},"company":{"n":32,"of":74}}},"percentiles":{"industry":{"views":0.3635972461273666,"likes":0.3261617900172117,"eng":0.34036144578313254,"save":0.33777969018932874},"company":{"views":0.060810810810810814,"likes":0.033783783783783786,"eng":0.5202702702702703,"save":0.5743243243243243}},"peers":{"industry":1162,"company":74,"companyName":"OpenAI"},"medians":{"industry":{"views":39945,"likes":289,"eng":0.007331788144862199,"save":0.0007642791742968587},"company":{"views":161517.5,"likes":714.5,"eng":0.005359315036129835,"save":0.00048697197603225794},"form":{"n":215,"views":44446,"eng":0.007500781331388686}},"ratios":{"vsCompanyViews":0.10304146609500518,"vsCompanyLikes":0.11756473058082575,"viewsPerFollower":0.03640137530401918,"viewsPerDay":30.476345116356455},"followers":457208,"reads":[{"at":"2026-10-12T15:45:44.121Z","views":16643,"likes":84,"reposts":6,"quotes":1,"replies":1,"saves":9,"source":"x-search-cache"},{"at":"2026-10-12T20:03:09.416Z","views":16643,"likes":84,"reposts":6,"quotes":1,"replies":1,"saves":9,"source":"x-search-cache"},{"at":"2026-10-12T20:46:03.444Z","views":16643,"likes":84,"reposts":6,"quotes":1,"replies":1,"saves":9,"source":"x-search-cache"}],"chartData":"/api/charts/5338e4aeb4b9.json","copy":{"en":{"why":[{"factor":"reach-vs-company","text":"Its 16.6K views sit at 0.10x OpenAI's 162K median, ranking #70 of 74 OpenAI posts."},{"factor":"views-per-follower","text":"At 0.036 views per follower against a 0.20 median, it lands in the 22nd percentile."},{"factor":"launch-day","text":"It was #2 of 2 OpenAI posts that day, behind a SWE-bench Verified post at 21.3K views."}],"verdict":"It sits in the Quiet quadrant, at #819 of 1,162 AI chart posts.","news":{"headline":"OpenAI posts internal instruction following eval bar chart","dek":"The benchmark bar chart drew 16.6K views and 84 likes, below the AI chart median on both.","what":"OpenAI posted a benchmark bar chart on April 14, 2025, covering internal instruction following eval accuracy. It was one of 2 OpenAI posts that day, alongside a SWE-bench Verified accuracy post.","numbers":"It reached 16.6K views, #740 of 1,162 AI chart posts, against a 39.9K median among AI chart posts. Its score of 35 ranks #819 of 1,162 AI chart posts and #71 of 74 OpenAI posts.","lesson":"A benchmark bar chart with 8 models shown drew a 0.054% save rate and 0.55% engagement rate. The chart shows printed values from 29% to 54%, and the post still landed below the AI chart median on views.","implication":"At #71 of 74 OpenAI posts and 0.10x OpenAI's 162K median views, this chart post sits near the bottom of the account's own range. The gap to the 39.9K AI chart median is 16.6K against 39.9K."}},"zh":{"why":[{"factor":"reach-vs-company","text":"浏览量1.7万，仅为OpenAI中位数16.2万的0.10倍，在74条帖子中排第70。"},{"factor":"views-per-follower","text":"每粉丝浏览量0.036，低于中位数0.20，处于第22百分位。"},{"factor":"launch-day","text":"发布当天有2条帖子，本条浏览量在2条中排第2。"}],"verdict":"安静象限，得分35，在1162条AI图表帖中排第819。","news":{"headline":"OpenAI图表帖浏览量垫底","dek":"1.7万浏览量，互动率0.55%，远低于中位数。","what":"OpenAI于2025年4月14日发布1张图表，展示8个模型的基准准确率。当天另有1条帖子，主题为SWE-bench Verified accuracy。","numbers":"浏览量1.7万，在1162条AI图表帖中排第740，低于中位数4万。点赞84，互动率0.55%。","lesson":"单图表、8模型对比的帖子，浏览量仅为中位数的0.10倍：简洁表格并未带来高传播。","implication":"在OpenAI的74条帖子中排第71，浏览量排名第70，与自身中位数16.2万差距明显。"}},"lane":"en:amd/DeepSeek-V4.1-Flash zh:amd/DeepSeek-V4-Flash","checkedAt":"2026-10-13T20:16:45.641Z"}}