3 ms·
The InstructGPT paper only explains the RLHF part of how ChatGPT works. There's reason to believe that isn't enough to achieve ChatGPT's performance and behavio
by maleldil 4y ago
The InstructGPT paper only explains the RLHF part of how ChatGPT works. There's reason to believe that isn't enough to achieve ChatGPT's performance and behaviour (e.g. [1]). There are other components that make ChatGPT more powerful, and OpenAPI is not being open about them.
[1] https://yaofu.notion.site/How-does-GPT-Obtain-its-Ability-Tracing-Emergent-Abilities-of-Language-Models-to-their-Sources-b9a57ac0fcf74f30a1ab9e3e36fa1dc1#3e764cf66cd84fd481fb166ec43188ca https://yaofu.notion.site/How-does-GPT-Obtain-its-Ability-Tr...
- nl 4y agoI don't think that conclusion is clear at all. Indeed your own link has things they thought were unclear struck and and parts of the InstructGPT paper that explains them inserted. They do have newer models that aren't generally available that are different though.