course/pr_1069/ko/_app/immutable/nodes/9.92af261d.js

import{s as X,n as I,o as J}from"../chunks/scheduler.37c15a92.js";import{S as K,i as V,g as r,s as l,r as G,A as W,h as m,f as n,c as s,j as O,u as S,x as v,k as Q,y as Z,a,v as j,d as z,t as A,w as R}from"../chunks/index.2bf4358c.js";import{Y as tt}from"../chunks/Youtube.1e50a667.js";import{C as et}from"../chunks/CourseFloatingBanner.9ff4c771.js";import{H as nt,E as at}from"../chunks/getInferenceSnippets.24b50994.js";function lt(U){let o,C,d,w,i,b,f,P,p,T,u,Y="디코더 모델(Decoder models)은 트랜스포머 모델의 디코더만 사용합니다. 각각의 단계마다, 어텐션 레이어는 주어진 단어에 대해 문장 내에서 해당 단어 앞에 위치한 단어들에 대해서만 액세스 할 수 있습니다. 이러한 모델을 <em>자동 회귀(auto-regressive) 모델</em>이라고 부릅니다.",L,c,B="디코더 모델의 사전 학습은 보통 문장 내 다음 단어 예측을 반복하는 방식으로 이루어집니다.",E,$,D="이러한 모델은 텍스트 생성에 특화되어 있습니다.",H,h,F="디코더 모델 계열의 대표 주자들은 다음과 같습니다:",M,_,N='<li><a href="https://huggingface.co/transformers/model_doc/ctrl.html" rel="nofollow">CTRL</a></li> <li><a href="https://huggingface.co/docs/transformers/model_doc/openai-gpt" rel="nofollow">GPT</a></li> <li><a href="https://huggingface.co/transformers/model_doc/gpt2.html" rel="nofollow">GPT-2</a></li> <li><a href="https://huggingface.co/transformers/model_doc/transformerxl.html" rel="nofollow">Transformer XL</a></li>',y,g,k,x,q;return i=new nt({props:{title:"디코더 모델",local:"디코더-모델",headingTag:"h1"}}),f=new et({props:{chapter:1,classNames:"absolute z-10 right-0 top-0"}}),p=new tt({props:{id:"d_ixlCubqQw"}}),g=new at({props:{source:"https://github.com/huggingface/course/blob/main/chapters/ko/chapter1/6.mdx"}}),{c(){o=r("meta"),C=l(),d=r("p"),w=l(),G(i.$$.fragment),b=l(),G(f.$$.fragment),P=l(),G(p.$$.fragment),T=l(),u=r("p"),u.innerHTML=Y,L=l(),c=r("p"),c.textContent=B,E=l(),$=r("p"),$.textContent=D,H=l(),h=r("p"),h.textContent=F,M=l(),_=r("ul"),_.innerHTML=N,y=l(),G(g.$$.fragment),k=l(),x=r("p"),this.h()},l(t){const e=W("svelte-u9bgzb",document.head);o=m(e,"META",{name:!0,content:!0}),e.forEach(n),C=s(t),d=m(t,"P",{}),O(d).forEach(n),w=s(t),S(i.$$.fragment,t),b=s(t),S(f.$$.fragment,t),P=s(t),S(p.$$.fragment,t),T=s(t),u=m(t,"P",{"data-svelte-h":!0}),v(u)!=="svelte-1p9xm7k"&&(u.innerHTML=Y),L=s(t),c=m(t,"P",{"data-svelte-h":!0}),v(c)!=="svelte-vpsok7"&&(c.textContent=B),E=s(t),$=m(t,"P",{"data-svelte-h":!0}),v($)!=="svelte-2694ur"&&($.textContent=D),H=s(t),h=m(t,"P",{"data-svelte-h":!0}),v(h)!=="svelte-ov0q2c"&&(h.textContent=F),M=s(t),_=m(t,"UL",{"data-svelte-h":!0}),v(_)!=="svelte-jabf5n"&&(_.innerHTML=N),y=s(t),S(g.$$.fragment,t),k=s(t),x=m(t,"P",{}),O(x).forEach(n),this.h()},h(){Q(o,"name","hf:doc:metadata"),Q(o,"content",st)},m(t,e){Z(document.head,o),a(t,C,e),a(t,d,e),a(t,w,e),j(i,t,e),a(t,b,e),j(f,t,e),a(t,P,e),j(p,t,e),a(t,T,e),a(t,u,e),a(t,L,e),a(t,c,e),a(t,E,e),a(t,$,e),a(t,H,e),a(t,h,e),a(t,M,e),a(t,_,e),a(t,y,e),j(g,t,e),a(t,k,e),a(t,x,e),q=!0},p:I,i(t){q||(z(i.$$.fragment,t),z(f.$$.fragment,t),z(p.$$.fragment,t),z(g.$$.fragment,t),q=!0)},o(t){A(i.$$.fragment,t),A(f.$$.fragment,t),A(p.$$.fragment,t),A(g.$$.fragment,t),q=!1},d(t){t&&(n(C),n(d),n(w),n(b),n(P),n(T),n(u),n(L),n(c),n(E),n($),n(H),n(h),n(M),n(_),n(y),n(k),n(x)),n(o),R(i,t),R(f,t),R(p,t),R(g,t)}}}const st='{"title":"디코더 모델","local":"디코더-모델","sections":[],"depth":1}';function ot(U){return J(()=>{new URLSearchParams(window.location.search).get("fw")}),[]}class ut extends K{constructor(o){super(),V(this,o,ot,lt,X,{})}}export{ut as component};