Back to search

Article

Upscaling A Smaller LLM to More Parameters via Manual Regressive Distillation

2024-07-23

Abstract excerpt

<title>Abstract</title> <p>The rapid expansion of data and the increasing complexity of linguistic tasks necessitate the development of more powerful language models capable of understanding and generating human language with greater accuracy. The novel approach of scaling GPT-Neo to 70 billion parameters through manual regressive distillation presents a significant advancement, leveraging a systematic knowledge...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work
bdccf5ed-d3c5-5eb3-afa5-2f934d74e0ae
DOI
10.21203/rs.3.rs-4775436/v1
Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest
Interactive article relationship graphSelect a related publication card to move it into the centre and load its closest explainable connections. Solid lines are source-backed structured connections. Dashed lines are semantic discovery signals and are not scientific evidence.
Upscaling A Smaller LLM to More Parameters via Manual Regressive DistillationDOI 10.21203/rs.3.rs-4775436/v1
Select a neighboring publication to make it the new centre.