Jump to content

File:Optimal ratio of training tokens to model parameters for Chinchilla scaling law.png

Page contents not supported in other languages.
This is a file from the Wikimedia Commons
From Wikipedia, the free encyclopedia

Original file (2,400 × 1,500 pixels, file size: 156 KB, MIME type: image/png)

Summary

Description
English: https://epochai.org/blog/chinchilla-scaling-a-replication-attempt

Original caption:

Figure 1: Optimal ratio of training tokens to model parameters using our estimates. Shaded regions represent 80% confidence intervals. While our estimates are consistent with the scaling policy used for Chinchilla, their estimates of their parametric model are not.
Date Apr 17, 2024
Source https://epochai.org/blog
Author Epoch AI

Licensing

w:en:Creative Commons
attribution
This file is licensed under the Creative Commons Attribution 4.0 International license.
You are free:
  • to share – to copy, distribute and transmit the work
  • to remix – to adapt the work
Under the following conditions:
  • attribution – You must give appropriate credit, provide a link to the license, and indicate if changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.

Captions

Optimal ratio of training tokens to model parameters for Chinchilla scaling law

Items portrayed in this file

depicts

File history

Click on a date/time to view the file as it appeared at that time.

Date/TimeThumbnailDimensionsUserComment
current23:10, 24 September 2024Thumbnail for version as of 23:10, 24 September 20242,400 × 1,500 (156 KB)Cosmia NebulaUploaded a work by Epoch AI from https://epochai.org/blog with UploadWizard

The following page uses this file: