{"provider_url":"https://hatena.blog","blog_title":"yousan\u306e\u30e1\u30e2","html":"<iframe src=\"https://hatenablog-parts.com/embed?url=https%3A%2F%2Fayousanz.hatenadiary.jp%2Fentry%2F2024%2F01%2F13%2F145715\" title=\"ahxt/LiteLlama-460M-1T\u3092\u52d5\u304b\u3059 - yousan\u306e\u30e1\u30e2\" class=\"embed-card embed-blogcard\" scrolling=\"no\" frameborder=\"0\" style=\"display: block; width: 100%; height: 190px; max-width: 500px; margin: 10px 0px;\"></iframe>","title":"ahxt/LiteLlama-460M-1T\u3092\u52d5\u304b\u3059","author_name":"ayousanz","description":"\u521d\u3081\u306b \u74b0\u5883 \u6e96\u5099 \u5b9f\u884c \u30e2\u30c7\u30eb\u306e\u30ed\u30fc\u30c9 \u30b5\u30f3\u30d7\u30eb\u30d7\u30ed\u30f3\u30d7\u30c8 \u307e\u3069\u30de\u30ae \u307e\u3069\u30de\u30aeQA \u521d\u3081\u306b With the recent release of #TinyLlama, SLMs have attracted a lot of attention. I re-released my previously trained SLM - LiteLlama under the MIT license, which has 460M parameters trained with 1T tokens. I hope to contribute a bit to the community.https\u2026","version":"1.0","provider_name":"Hatena Blog","url":"https://ayousanz.hatenadiary.jp/entry/2024/01/13/145715","type":"rich","author_url":"https://blog.hatena.ne.jp/ayousanz/","published":"2024-01-13 14:57:15","blog_url":"https://ayousanz.hatenadiary.jp/","categories":["AI"],"width":"100%","image_url":null,"height":"190"}