“What we have been building since ChatGPT at H4.

• No pretraining in any way

Basic Three Steps

Goal: “helpful, harmless, honest, and huggy” bots.

1. Retraining step: large-scale next token prediction
2. Incontext learning: few shot learning without updating parameters
3. “Helpful” steps
   1. Taking supervised data to perform supervised fine tuning
4. “Harmless” steps
   1. Training a classifier for result ranking
   2. RLHF

Benchmarking

Before we started to train, we have a problem. Most benchmarks are on generic reasoning, which evaluates 1), 2). Therefore, we need new metrics for steps 4) and 5).

the transformational generative syntax is a linguistical precept proposed by Noam Chomsky which has the interesting conclusion that meaning is supported by structure, rather than the other way around as generative semantics suggests.

This means that you can first come up with generic, independent structure to a sentence, then fill in the sentence with meaning.

For instance, “colorless green ideas sleep furiously” is a sentence Noam Chomsky proposes to have perfect structure but failes to be filled with meaning, supporting the transformational generative syntax theory.

Translation Theory

Translation Theory is the theory that studies how translation works.

Spectrum of Translation

domestication and foreignization are processes by which a translator can choose to alter the style of a translation for a purpose.

foreignization

trying to bring the target language closer to the source language

• bring in foreign words
• use colourful idioms
• use old words

domestication

trying to bring he source language closer to the target language