github
/
minimind-v
mirror of https://github.com/jingyaogong/minimind-v.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

update readme

pull/9/head
gongjy 2024-10-05 00:50:26 +08:00
parent b671c1a2be
commit aadb17350f
2 changed files with 8 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
README.md
View File

@ -219,11 +219,15 @@ MiniMind-V的结构几乎不变，仅增加Visual Encoder和特征投影两个
机器并不会区分这些信号的模态来源，而只是处理和分析这些序列背后所承载的信息内容。 机器并不会区分这些信号的模态来源，而只是处理和分析这些序列背后所承载的信息内容。
--- ---
私以为，**G**enerative **P**retrained **T**ransformer (GPT) 比 **L**arge **L**anguage **M**odel (LLM)更为贴切， 私以为，**G**enerative **P**retrained **T**ransformer (GPT) 比 **L**arge **L**anguage **M**odel (LLM)更为贴切，
因此本人表达上更习惯用"GPT"去代表LLM/VLM/类GPT架构的系列模型，而非为了蹭OpenAI的热度。 因此本人表达上更习惯用"GPT"去代表LLM/VLM/类GPT架构的系列模型，而非为了蹭OpenAI的热度。
--- ---
至此，我们可以用一句话总结GPT的所作所为： 至此，我们可以用一句话总结GPT的所作所为：
GPT模型根据现有token预测输出下一个下下一个下下下一个token ...，直到模型输出结束符；此处的"token"其实并不需要一定是文本！ GPT模型根据现有token预测输出下一个下下一个下下下一个token ...，直到模型输出结束符；此处的"token"其实并不需要一定是文本！
--- ---
* 对于LLM模型，如果需要理解"图片"，我们只要把"图片"作为对一种特殊的从来没见过的"外国语言"，通过"外语词典"翻译后即可作为特殊的语言输入LLM * 对于LLM模型，如果需要理解"图片"，我们只要把"图片"作为对一种特殊的从来没见过的"外国语言"，通过"外语词典"翻译后即可作为特殊的语言输入LLM

4
README_en.md
View File

@ -207,11 +207,15 @@ At this point, it's interesting to ponder two questions: What is a **L**arge **L
Machines do not differentiate the modality source of these signals but rather process and analyze the information content carried by these sequences. Machines do not differentiate the modality source of these signals but rather process and analyze the information content carried by these sequences.
--- ---
I personally believe that **G**enerative **P**retrained **T**ransformer (GPT) is a more fitting term than **L**arge **L**anguage **M**odel (LLM), I personally believe that **G**enerative **P**retrained **T**ransformer (GPT) is a more fitting term than **L**arge **L**anguage **M**odel (LLM),
and thus I prefer to use "GPT" to represent LLM/VLM/GPT-like architectures, rather than to piggyback on OpenAI's popularity. and thus I prefer to use "GPT" to represent LLM/VLM/GPT-like architectures, rather than to piggyback on OpenAI's popularity.
--- ---
In summary, we can encapsulate what GPT does in one sentence: In summary, we can encapsulate what GPT does in one sentence:
GPT models predict the next, and the next, and the next token... until the model outputs an end token; here, the "token" does not necessarily have to be text! GPT models predict the next, and the next, and the next token... until the model outputs an end token; here, the "token" does not necessarily have to be text!
--- ---
* For LLM models, if understanding "images" is required, we can treat "images" as a special kind of "foreign language" never seen before, translating them through a "foreign dictionary" into a special language input for the LLM. * For LLM models, if understanding "images" is required, we can treat "images" as a special kind of "foreign language" never seen before, translating them through a "foreign dictionary" into a special language input for the LLM.