Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Jan 8, 2023

• Update README.md

  Show description for 37baab7
  karpathy

  authored
  37baab7

Commits on Aug 4, 2022

• Merge pull request #84 from ericjang/master

  Show description for 7218bcf
  karpathy

  authored
  7218bcf

Commits on Aug 3, 2022

• Add setup.py to allow mingpt to be used as a third-party library

  ericjang

  committed
  48c815b

Commits on Jul 29, 2022

• Merge pull request #83 from mishig25/patch-1

  Show description for cafce45
  karpathy

  authored
  cafce45

Commits on Jul 28, 2022

• Use XOR operator ^ for checking assertion type_given XOR params_given

  Show description for 90420ee
  mishig25

  authored
  90420ee

Commits on Jul 27, 2022

• Merge pull request #82 from neverix/patch-1

  Show description for ca74e9a
  karpathy

  authored
  ca74e9a

• Fix README.md typo

  Show description for e461bf6
  neverix

  authored
  e461bf6

Commits on Jul 26, 2022

• Merge pull request #81 from luigidisotto/callbacks-optimizer

  Show description for 31559f7
  karpathy

  authored
  31559f7

• Add optimizer to Trainer's self for callbacks.

  luigidisotto

  committed
  c4c650e

Commits on Jul 12, 2022

• use a bit more extended example that has my last name too because nice to show how it breaks up into more tokens

  karpathy

  authored
  e2065c5

• add a full example into the script as well

  karpathy

  authored
  d8dd157

• Merge pull request #78 from nat/patch-1

  Show description for 59fea1b
  karpathy

  authored
  59fea1b

• Typos

  Show description for e9f6e3d
  nat

  authored
  e9f6e3d

• adjust the readme docs to reflect bpe changes

  karpathy

  authored
  0fc12d7

• add a refactored BPE encoder from openai. Basically I dont super trust the huggingface tokenizer, the implementation sprawls multiple files and inheritance and has special magic handling around Add…

  Show description for 9642f40
  karpathy

  authored
  9642f40

Commits on Jul 11, 2022

• few added todo notes to readme

  karpathy

  authored
  40635a9

• refactor sequence generation into the model and match the huggingface/transformers api. touches everything but this makes a lot more sense to me aesthetically

  karpathy

  authored
  acaadac

Commits on Jul 9, 2022

• small typo fixes in readme

  karpathy

  authored
  5af9e5c

• quick docs on some planned todos

  karpathy

  authored
  610baf2

Commits on Jul 8, 2022

• make generation script into a notebook, makes much more sense that way i think, and much easier to use

  karpathy

  authored
  12f346a

• rename the script to make sense

  karpathy

  authored
  b14c991

• refactor pretrained weight loading into from_pretrained and add unit tests

  karpathy

  authored
  803f388

• fix parameter counting

  karpathy

  authored
  4a56b20

• ocd is killing me

  karpathy

  authored
  28bcbd0

• updates to readme

  karpathy

  authored
  449a980

• remove legacy notebooks

  karpathy

  authored
  b7c9acc

• refactor readme to match the repo

  karpathy

  authored
  e7fe548

• simple notebook demo showing how to use minGPT

  karpathy

  committed
  7569ab9

Commits on Jul 1, 2022

• ummm eyeroll

  karpathy

  authored
  2e979dd

• split out register_callback to set/add

  karpathy

  committed
  2f3400f

• add maxiters to trainer

  karpathy

  committed
  d9ea878

Commits on Jun 29, 2022

• changing default seed to the suspect SOTA value as per https://arxiv.org/abs/2109.08203

  karpathy

  committed
  1f33571

Commits on Jun 27, 2022

• no

  karpathy

  authored
  7f6775a

• ok i hated the previous global/local config idea. reverting it and simplying and i think this is the best api so far

  karpathy

  authored
  00aa9cb

• be more defensive around model_type, don't let the user shoot themselves in the foot

  karpathy

  authored
  ea20661

Pagination

PreviousNext