English
English Indonesia Türkçe Deutsch Español Português Pусский Français Italiano Nederlands Polski العربية ไทย Tiếng Việt 繁體中文 简体中文 日本語 한국어
sokeyer's profile picture. Developer, enjoy the life! Machine learning and Large Language Models

Xin Liu

@sokeyer

Developer, enjoy the life! Machine learning and Large Language Models

7KPosts 58Followers 624Following
You might like
Xin Liu reposted
karpathy's profile picture.
Andrej Karpathy
@karpathy
Feb 5

New 3h31m video on YouTube: "Deep Dive into LLMs like ChatGPT" This is a general audience deep dive into the Large Language Model (LLM) AI technology that powers ChatGPT and related products. It is covers the full training stack of how the models are developed, along with mental…

karpathy's tweet image. New 3h31m video on YouTube: "Deep Dive into LLMs like ChatGPT" This is a general audience deep dive into the Large Language Model (LLM) AI technology that powers ChatGPT and related products. It is covers the full training stack of how the models are developed, along with mental…
778
3K
20K
15K
2.4M
sokeyer's profile picture.
Xin Liu
 @sokeyer
Sep 21, 2024

Why I Won't Use Next.js, article epicweb.dev/why-i-wont-use…

0
0
1
0
57
Xin Liu reposted
karpathy's profile picture.
Andrej Karpathy
@karpathy
Dec 11, 2023

Official post on Mixtral 8x7B: mistral.ai/news/mixtral-o… Official PR into vLLM shows the inference code: github.com/vllm-project/v… New HuggingFace explainer on MoE very nice: huggingface.co/blog/moe In naive decoding, performance of a bit above 70B (Llama 2), at inference speed…

karpathy's tweet card. Mixture of Experts Explained
Mixture of Experts Explained
Source: huggingface.co
GuillaumeLample's profile picture. Cofounder & Chief Scientist https://t.co/hLfvKLldwL (@MistralAI). Working on LLMs. Ex @MetaAI | PhD @Sorbonne_Univ_ | MSc @CarnegieMellon | X11 @Polytechnique
Guillaume Lample @ NeurIPS 2024
@GuillaumeLample
Dec 11, 2023

Very excited to release our second model, Mixtral 8x7B, an open weight mixture of experts model. Mixtral matches or outperforms Llama 2 70B and GPT3.5 on most benchmarks, and has the inference speed of a 12B dense model. It supports a context length of 32k tokens. (1/n)

GuillaumeLample's tweet image. Very excited to release our second model, Mixtral 8x7B, an open weight mixture of experts model. Mixtral matches or outperforms Llama 2 70B and GPT3.5 on most benchmarks, and has the inference speed of a 12B dense model. It supports a context length of 32k tokens. (1/n)
GuillaumeLample's tweet image. Very excited to release our second model, Mixtral 8x7B, an open weight mixture of experts model. Mixtral matches or outperforms Llama 2 70B and GPT3.5 on most benchmarks, and has the inference speed of a 12B dense model. It supports a context length of 32k tokens. (1/n)
GuillaumeLample's tweet image. Very excited to release our second model, Mixtral 8x7B, an open weight mixture of experts model. Mixtral matches or outperforms Llama 2 70B and GPT3.5 on most benchmarks, and has the inference speed of a 12B dense model. It supports a context length of 32k tokens. (1/n)
80
570
4K
1K
2.2M
56
581
3K
2K
681K
sokeyer's profile picture.
Xin Liu
 @sokeyer
Jul 6, 2023

I just got 830 points✨ on nextjs.org/learn 来自 @vercel

0
0
0
0
30
sokeyer's profile picture.
Xin Liu
 @sokeyer
Jun 29, 2023

Sometimes, I foget why I do it.

0
0
2
0
20
sokeyer's profile picture.
Xin Liu
 @sokeyer
Jun 11, 2023

还记得你何时加入 Twitter 吗？我知道！#我的Twitter周年纪念日

sokeyer's tweet image. 还记得你何时加入 Twitter 吗？我知道！#我的Twitter周年纪念日
0
1
3
0
27
sokeyer's profile picture.
Xin Liu
 @sokeyer
Feb 24, 2023

Back to China. Everything fine.

0
0
2
0
61
sokeyer's profile picture.
Xin Liu
 @sokeyer
Feb 14, 2023

在东京的第21天。

0
0
0
0
55

United States Trends

You might like

Something went wrong.


Something went wrong.