{
 "<h1>GPT-NeoX Checkpoints</h1>\n": "<h1>GPT-\u30cd\u30aa\u30c3\u30af\u30b9\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8</h1>\n",
 "<h2>Download all checkpoint files</h2>\n": "<h2>\u3059\u3079\u3066\u306e\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8\u30d5\u30a1\u30a4\u30eb\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9</h2>\n",
 "<h3>Get files to download</h3>\n<ul><p><em>Returns</em>  a list of files to be downloaded</p></ul>\n": "<h3>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u30d5\u30a1\u30a4\u30eb\u3092\u53d6\u5f97</h3>\n<ul><p><em>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u30d5\u30a1\u30a4\u30eb\u306e\u30ea\u30b9\u30c8\u3092\u8fd4\u3057\u307e\u3059</em></p></ul>\n",
 "<h3>Load a pair of checkpoint files</h3>\n<ul><li><span translate=no>_^_0_^_</span>  pair of files to load </li>\n<p><em>Returns</em>  the loaded parameter tensors</p></ul>\n": "<h3>1 \u7d44\u306e\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8\u30d5\u30a1\u30a4\u30eb\u3092\u30ed\u30fc\u30c9\u3059\u308b</h3>\n<ul><li><span translate=no>_^_0_^_</span>\u30ed\u30fc\u30c9\u3059\u308b\u30d5\u30a1\u30a4\u30eb\u306e\u30da\u30a2</li>\n</ul><p><em>\u30ed\u30fc\u30c9\u3055\u308c\u305f\u30d1\u30e9\u30e1\u30fc\u30bf\u30c6\u30f3\u30bd\u30eb\u3092\u8fd4\u3057\u307e\u3059</em>\u3002</p>\n",
 "<h3>Load a parameter by merging the partitions along first dimension</h3>\n<ul><li><span translate=no>_^_0_^_</span>  is the parameter </li>\n<li><span translate=no>_^_1_^_</span>  is the name of the parameter </li>\n<li><span translate=no>_^_2_^_</span>  first partition dictionary </li>\n<li><span translate=no>_^_3_^_</span>  second partition dictionary</li></ul>\n": "<h3>1 \u756a\u76ee\u306e\u6b21\u5143\u306b\u6cbf\u3063\u3066\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u3092\u30de\u30fc\u30b8\u3057\u3066\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u30ed\u30fc\u30c9\u3057\u307e\u3059</h3>\n<ul><li><span translate=no>_^_0_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u540d\u524d\u3067\u3059</li>\n<li><span translate=no>_^_2_^_</span>\u7b2c 1 \u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li>\n<li><span translate=no>_^_3_^_</span>2 \u756a\u76ee\u306e\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li></ul>\n",
 "<h3>Load a parameter by merging the partitions along second dimension</h3>\n<ul><li><span translate=no>_^_0_^_</span>  is the parameter </li>\n<li><span translate=no>_^_1_^_</span>  is the name of the parameter </li>\n<li><span translate=no>_^_2_^_</span>  first partition dictionary </li>\n<li><span translate=no>_^_3_^_</span>  second partition dictionary</li></ul>\n": "<h3>2 \u756a\u76ee\u306e\u6b21\u5143\u306b\u6cbf\u3063\u3066\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u3092\u30de\u30fc\u30b8\u3057\u3066\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u30ed\u30fc\u30c9\u3057\u307e\u3059</h3>\n<ul><li><span translate=no>_^_0_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u540d\u524d\u3067\u3059</li>\n<li><span translate=no>_^_2_^_</span>\u7b2c 1 \u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li>\n<li><span translate=no>_^_3_^_</span>2 \u756a\u76ee\u306e\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li></ul>\n",
 "<h3>Load an un-partitioned parameter</h3>\n<p>This does a sanity check to make use both partitions are the same</p>\n<ul><li><span translate=no>_^_0_^_</span>  is the parameter </li>\n<li><span translate=no>_^_1_^_</span>  is the name of the parameter </li>\n<li><span translate=no>_^_2_^_</span>  first partition dictionary </li>\n<li><span translate=no>_^_3_^_</span>  second partition dictionary</li></ul>\n": "<h3>\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u5316\u3055\u308c\u3066\u3044\u306a\u3044\u30d1\u30e9\u30e1\u30fc\u30bf\u3092\u30ed\u30fc\u30c9\u3059\u308b</h3>\n<p>\u3053\u308c\u306b\u3088\u308a\u3001\u4e21\u65b9\u306e\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u304c\u540c\u3058\u3067\u3042\u308b\u3053\u3068\u3092\u78ba\u8a8d\u3059\u308b\u305f\u3081\u306e\u30b5\u30cb\u30c6\u30a3\u30c1\u30a7\u30c3\u30af\u304c\u884c\u308f\u308c\u307e\u3059\u3002</p>\n<ul><li><span translate=no>_^_0_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u540d\u524d\u3067\u3059</li>\n<li><span translate=no>_^_2_^_</span>\u7b2c 1 \u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li>\n<li><span translate=no>_^_3_^_</span>2 \u756a\u76ee\u306e\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li></ul>\n",
 "<h3>Load biases that are partitioned which gets added on reduce</h3>\n<ul><li><span translate=no>_^_0_^_</span>  is the parameter </li>\n<li><span translate=no>_^_1_^_</span>  is the name of the parameter </li>\n<li><span translate=no>_^_2_^_</span>  first partition dictionary </li>\n<li><span translate=no>_^_3_^_</span>  second partition dictionary</li></ul>\n": "<h3>\u5206\u5272\u3055\u308c\u305f\u8ca0\u8377\u30d0\u30a4\u30a2\u30b9\u304c\u30ea\u30c7\u30e5\u30fc\u30b9\u6642\u306b\u8ffd\u52a0\u3055\u308c\u308b</h3>\n<ul><li><span translate=no>_^_0_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u3067\u3059</li>\n<li><span translate=no>_^_1_^_</span>\u306f\u30d1\u30e9\u30e1\u30fc\u30bf\u306e\u540d\u524d\u3067\u3059</li>\n<li><span translate=no>_^_2_^_</span>\u7b2c 1 \u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li>\n<li><span translate=no>_^_3_^_</span>2 \u756a\u76ee\u306e\u30d1\u30fc\u30c6\u30a3\u30b7\u30e7\u30f3\u8f9e\u66f8</li></ul>\n",
 "<p> </p>\n": "<p></p>\n",
 "<p>Download </p>\n": "<p>[\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9]</p>\n",
 "<p>Download path </p>\n": "<p>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u30d1\u30b9</p>\n",
 "<p>Embedding layer </p>\n": "<p>\u57cb\u3081\u8fbc\u307f\u30ec\u30a4\u30e4\u30fc</p>\n",
 "<p>Empty states (not used) </p>\n": "<p>\u7a7a\u306e\u72b6\u614b (\u672a\u4f7f\u7528)</p>\n",
 "<p>Final normalization layer and readout layer </p>\n": "<p>\u6700\u7d42\u6b63\u898f\u5316\u5c64\u3068\u8aad\u307f\u51fa\u3057\u5c64</p>\n",
 "<p>Get files to download </p>\n": "<p>\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u30d5\u30a1\u30a4\u30eb\u3092\u53d6\u5f97</p>\n",
 "<p>Iterate </p>\n": "<p>\u7e70\u308a\u8fd4\u3057</p>\n",
 "<p>Layer checkpoints </p>\n": "<p>\u30ec\u30a4\u30e4\u30fc\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8</p>\n",
 "<p>Log </p>\n": "<p>\u30ed\u30b0</p>\n",
 "<p>Parent url </p>\n": "<p>\u4fdd\u8b77\u8005\u306e URL</p>\n",
 "<p>Transformer layers </p>\n": "<p>\u30c8\u30e9\u30f3\u30b9\u30d5\u30a9\u30fc\u30de\u30fc\u5c64</p>\n",
 "<p>Vocabulary and configs </p>\n": "<p>\u8a9e\u5f59\u3068\u69cb\u6210</p>\n",
 "Code to download checkpoints and helpers to load them.": "\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8\u3068\u305d\u308c\u3089\u3092\u30ed\u30fc\u30c9\u3059\u308b\u30d8\u30eb\u30d1\u30fc\u3092\u30c0\u30a6\u30f3\u30ed\u30fc\u30c9\u3059\u308b\u30b3\u30fc\u30c9\u3002",
 "GPT-NeoX Checkpoints": "GPT-\u30cd\u30aa\u30c3\u30af\u30b9\u30c1\u30a7\u30c3\u30af\u30dd\u30a4\u30f3\u30c8"
}