ml-summit / src /figures /gpt4.html
facat's picture
init
2fc4496 unverified
raw
history blame
No virus
19.6 kB
<div data-v-22aefd69="" class="observablehq"><svg width="900" height="625" viewBox="0,0,900,625" style="max-width: 100%; height: auto; height: intrinsic;"><g transform="translate(0,505)" fill="none" font-size="10" font-family="sans-serif" text-anchor="middle"><path class="domain" stroke="currentColor" d="M34,0H860"></path><g class="tick" opacity="1" transform="translate(52.91603053435112,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Calculus BC</text></g><g class="tick" opacity="1" transform="translate(84.44274809160304,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AMC 12</text></g><g class="tick" opacity="1" transform="translate(115.96946564885495,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">Codeforces Rating</text></g><g class="tick" opacity="1" transform="translate(147.49618320610688,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP English Literature</text></g><g class="tick" opacity="1" transform="translate(179.0229007633588,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AMC 10</text></g><g class="tick" opacity="1" transform="translate(210.54961832061068,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">Uniform Bar Exam</text></g><g class="tick" opacity="1" transform="translate(242.0763358778626,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP English Language</text></g><g class="tick" opacity="1" transform="translate(273.60305343511453,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Chemistry</text></g><g class="tick" opacity="1" transform="translate(305.12977099236645,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">GRE Quantitative</text></g><g class="tick" opacity="1" transform="translate(336.6564885496183,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Physics 2</text></g><g class="tick" opacity="1" transform="translate(368.1832061068702,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">USABO Semifinal 2020</text></g><g class="tick" opacity="1" transform="translate(399.70992366412213,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Macroeconomics</text></g><g class="tick" opacity="1" transform="translate(431.23664122137404,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Statistics</text></g><g class="tick" opacity="1" transform="translate(462.76335877862596,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">LSAT</text></g><g class="tick" opacity="1" transform="translate(494.29007633587787,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">GRE Writing</text></g><g class="tick" opacity="1" transform="translate(525.8167938931297,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Microeconomics</text></g><g class="tick" opacity="1" transform="translate(557.3435114503816,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Biology</text></g><g class="tick" opacity="1" transform="translate(588.8702290076335,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">GRE Verbal</text></g><g class="tick" opacity="1" transform="translate(620.3969465648854,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP World History</text></g><g class="tick" opacity="1" transform="translate(651.9236641221373,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">SAT Math</text></g><g class="tick" opacity="1" transform="translate(683.4503816793892,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP US History</text></g><g class="tick" opacity="1" transform="translate(714.9770992366412,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP US Government</text></g><g class="tick" opacity="1" transform="translate(746.5038167938931,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Psychology</text></g><g class="tick" opacity="1" transform="translate(778.030534351145,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Art History</text></g><g class="tick" opacity="1" transform="translate(809.5572519083969,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">SAT EBRW</text></g><g class="tick" opacity="1" transform="translate(841.0839694656488,0)"><line stroke="currentColor" y2="6"></line><text fill="currentColor" y="9" dy=".25em" dx="0.4em" transform="rotate(70)" style="text-anchor: start; font-size: 10px;">AP Environmental Science</text></g></g><g transform="translate(0,505)"><text fill="black" x="50%" y="115" text-anchor="middle" font-family="sans-serif" font-size="10px">Exam</text></g><g transform="translate(34,0)" class="axisLeft" fill="none" font-size="10" font-family="sans-serif" text-anchor="end"><path class="domain" stroke="currentColor" d="M-6,505H0V55H-6" stroke-opacity="0"></path><g class="tick" opacity="1" transform="translate(0,505)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">0%</text></g><g class="tick" opacity="1" transform="translate(0,415)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">20%</text></g><g class="tick" opacity="1" transform="translate(0,325)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">40%</text></g><g class="tick" opacity="1" transform="translate(0,235)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">60%</text></g><g class="tick" opacity="1" transform="translate(0,144.99999999999997)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">80%</text></g><g class="tick" opacity="1" transform="translate(0,55)"><line stroke="currentColor" x2="-6"></line><text fill="currentColor" x="-9" dy="0.32em">100%</text></g></g><g transform="translate(34,0)"><text fill="black" x="-34" y="41" text-anchor="start" font-family="sans-serif" font-size="10px">Estimated percentile lower bound (among test takers)</text></g><line class="gridline" x1="34" x2="860" y1="505" y2="505" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><line class="gridline" x1="34" x2="860" y1="415" y2="415" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><line class="gridline" x1="34" x2="860" y1="325" y2="325" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><line class="gridline" x1="34" x2="860" y1="235" y2="235" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><line class="gridline" x1="34" x2="860" y1="144.99999999999997" y2="144.99999999999997" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><line class="gridline" x1="34" x2="860" y1="55" y2="55" stroke="#C5C5D2" stroke-width="1" opacity="0.2"></line><text x="0" y="20" text-anchor="left" class="chart-title" font-family="sans-serif" font-weight="bold" font-size="14px">Exam results (ordered by GPT-3.5 performance)</text><rect class="bar-gpt4" x="40.30534351145036" y="311.5" width="25.22137404580153" height="193.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="71.83206106870227" y="302.5" width="25.22137404580153" height="202.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="103.35877862595419" y="482.5" width="25.22137404580153" height="22.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="134.8854961832061" y="469" width="25.22137404580153" height="36" fill="#51DA4C"></rect><rect class="bar-gpt4" x="166.41221374045801" y="478" width="25.22137404580153" height="27" fill="#51DA4C"></rect><rect class="bar-gpt4" x="197.9389312977099" y="99.99999999999999" width="25.22137404580153" height="405" fill="#51DA4C"></rect><rect class="bar-gpt4" x="229.4656488549618" y="442" width="25.22137404580153" height="63" fill="#51DA4C"></rect><rect class="bar-gpt4" x="260.99236641221376" y="185.5" width="25.22137404580153" height="319.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="292.51908396946567" y="144.99999999999997" width="25.22137404580153" height="360" fill="#51DA4C"></rect><rect class="bar-gpt4" x="324.0458015267175" y="208" width="25.22137404580153" height="297" fill="#51DA4C"></rect><rect class="bar-gpt4" x="355.57251908396944" y="59.50000000000001" width="25.22137404580153" height="445.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="387.09923664122135" y="127" width="25.22137404580153" height="378" fill="#51DA4C"></rect><rect class="bar-gpt4" x="418.62595419847327" y="122.50000000000001" width="25.22137404580153" height="382.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="450.1526717557252" y="109" width="25.22137404580153" height="396" fill="#51DA4C"></rect><rect class="bar-gpt4" x="481.6793893129771" y="262" width="25.22137404580153" height="243" fill="#51DA4C"></rect><rect class="bar-gpt4" x="513.206106870229" y="136" width="25.22137404580153" height="369" fill="#51DA4C"></rect><rect class="bar-gpt4" x="544.7328244274809" y="122.50000000000001" width="25.22137404580153" height="382.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="576.2595419847328" y="59.50000000000001" width="25.22137404580153" height="445.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="607.7862595419847" y="212.5" width="25.22137404580153" height="292.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="639.3129770992366" y="104.5" width="25.22137404580153" height="400.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="670.8396946564885" y="104.5" width="25.22137404580153" height="400.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="702.3664122137404" y="109" width="25.22137404580153" height="396" fill="#51DA4C"></rect><rect class="bar-gpt4" x="733.8931297709923" y="131.50000000000003" width="25.22137404580153" height="373.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="765.4198473282443" y="118" width="25.22137404580153" height="387" fill="#51DA4C"></rect><rect class="bar-gpt4" x="796.9465648854962" y="86.49999999999997" width="25.22137404580153" height="418.5" fill="#51DA4C"></rect><rect class="bar-gpt4" x="828.4732824427481" y="95.49999999999999" width="25.22137404580153" height="409.5" fill="#51DA4C"></rect><rect class="bar-gpt4novision" x="40.30534351145036" y="311.5" width="25.22137404580153" height="193.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="71.83206106870227" y="419.5" width="25.22137404580153" height="85.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="103.35877862595419" y="482.5" width="25.22137404580153" height="22.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="134.8854961832061" y="469" width="25.22137404580153" height="36" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="166.41221374045801" y="460" width="25.22137404580153" height="45" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="197.9389312977099" y="99.99999999999999" width="25.22137404580153" height="405" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="229.4656488549618" y="442" width="25.22137404580153" height="63" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="260.99236641221376" y="185.5" width="25.22137404580153" height="319.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="292.51908396946567" y="226" width="25.22137404580153" height="279" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="324.0458015267175" y="208" width="25.22137404580153" height="297" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="355.57251908396944" y="59.50000000000001" width="25.22137404580153" height="445.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="387.09923664122135" y="127" width="25.22137404580153" height="378" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="418.62595419847327" y="122.50000000000001" width="25.22137404580153" height="382.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="450.1526717557252" y="131.50000000000003" width="25.22137404580153" height="373.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="481.6793893129771" y="262" width="25.22137404580153" height="243" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="513.206106870229" y="235" width="25.22137404580153" height="270" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="544.7328244274809" y="122.50000000000001" width="25.22137404580153" height="382.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="576.2595419847328" y="73.00000000000001" width="25.22137404580153" height="432" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="607.7862595419847" y="212.5" width="25.22137404580153" height="292.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="639.3129770992366" y="104.5" width="25.22137404580153" height="400.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="670.8396946564885" y="172" width="25.22137404580153" height="333" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="702.3664122137404" y="109" width="25.22137404580153" height="396" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="733.8931297709923" y="131.50000000000003" width="25.22137404580153" height="373.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="765.4198473282443" y="118" width="25.22137404580153" height="387" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="796.9465648854962" y="86.49999999999997" width="25.22137404580153" height="418.5" fill="#7CF178"></rect><rect class="bar-gpt4novision" x="828.4732824427481" y="95.49999999999999" width="25.22137404580153" height="409.5" fill="#7CF178"></rect><rect class="bar-gpt35" x="40.30534351145036" y="505" width="25.22137404580153" height="0" fill="#3C46FF"></rect><rect class="bar-gpt35" x="71.83206106870227" y="486.99999999999994" width="25.22137404580153" height="18.000000000000057" fill="#3C46FF"></rect><rect class="bar-gpt35" x="103.35877862595419" y="482.5" width="25.22137404580153" height="22.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="134.8854961832061" y="469" width="25.22137404580153" height="36" fill="#3C46FF"></rect><rect class="bar-gpt35" x="166.41221374045801" y="460" width="25.22137404580153" height="45" fill="#3C46FF"></rect><rect class="bar-gpt35" x="197.9389312977099" y="460" width="25.22137404580153" height="45" fill="#3C46FF"></rect><rect class="bar-gpt35" x="229.4656488549618" y="442" width="25.22137404580153" height="63" fill="#3C46FF"></rect><rect class="bar-gpt35" x="260.99236641221376" y="406.00000000000006" width="25.22137404580153" height="98.99999999999994" fill="#3C46FF"></rect><rect class="bar-gpt35" x="292.51908396946567" y="392.5" width="25.22137404580153" height="112.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="324.0458015267175" y="370" width="25.22137404580153" height="135" fill="#3C46FF"></rect><rect class="bar-gpt35" x="355.57251908396944" y="365.5" width="25.22137404580153" height="139.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="387.09923664122135" y="356.49999999999994" width="25.22137404580153" height="148.50000000000006" fill="#3C46FF"></rect><rect class="bar-gpt35" x="418.62595419847327" y="325" width="25.22137404580153" height="180" fill="#3C46FF"></rect><rect class="bar-gpt35" x="450.1526717557252" y="325" width="25.22137404580153" height="180" fill="#3C46FF"></rect><rect class="bar-gpt35" x="481.6793893129771" y="262" width="25.22137404580153" height="243" fill="#3C46FF"></rect><rect class="bar-gpt35" x="513.206106870229" y="235" width="25.22137404580153" height="270" fill="#3C46FF"></rect><rect class="bar-gpt35" x="544.7328244274809" y="226" width="25.22137404580153" height="279" fill="#3C46FF"></rect><rect class="bar-gpt35" x="576.2595419847328" y="221.5" width="25.22137404580153" height="283.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="607.7862595419847" y="212.5" width="25.22137404580153" height="292.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="639.3129770992366" y="190.00000000000003" width="25.22137404580153" height="315" fill="#3C46FF"></rect><rect class="bar-gpt35" x="670.8396946564885" y="172" width="25.22137404580153" height="333" fill="#3C46FF"></rect><rect class="bar-gpt35" x="702.3664122137404" y="158.5" width="25.22137404580153" height="346.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="733.8931297709923" y="131.50000000000003" width="25.22137404580153" height="373.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="765.4198473282443" y="118" width="25.22137404580153" height="387" fill="#3C46FF"></rect><rect class="bar-gpt35" x="796.9465648854962" y="113.5" width="25.22137404580153" height="391.5" fill="#3C46FF"></rect><rect class="bar-gpt35" x="828.4732824427481" y="95.49999999999999" width="25.22137404580153" height="409.5" fill="#3C46FF"></rect><g transform="translate(855, 16)" text-anchor="end" font-family="sans-serif" font-size="12"><g transform="translate(0,0)"><rect x="-10" width="10" height="10" fill="#51DA4C"></rect><text x="-16" y="5" dy="0.35em">gpt-4</text></g><g transform="translate(0,12)"><rect x="-10" width="10" height="10" fill="#7CF178"></rect><text x="-16" y="5" dy="0.35em">gpt-4 (no vision)</text></g><g transform="translate(0,24)"><rect x="-10" width="10" height="10" fill="#3C46FF"></rect><text x="-16" y="5" dy="0.35em">gpt3.5</text></g></g></svg></div>