Commits

Branch selector

 master

User selector

All users

Datepicker

All time

Commit History

Commits on Aug 17, 2023

• calculate the freq_cis online, no need to write/read them to/from checkpoints

  karpathy

  committed
  bd18228

  Copy full SHA for bd18228

• Merge pull request karpathy#307 from madroidmaq/master

  Show description for b68a6d2
  karpathy

  authored
  b68a6d2

  Copy full SHA for b68a6d2

Commits on Aug 16, 2023

• Merge pull request karpathy#306 from rdentato/patch-utf8-no-validation

  Show description for 57bf0e9
  karpathy

  authored
  57bf0e9

  Copy full SHA for 57bf0e9

• Jupter Notebook: Add run Meta's Llama 2 models

  madroidmaq

  committed
  9fbe96f

  Copy full SHA for 9fbe96f

• Added space to str_buffer in case max_token_length is 1.

  rdentato

  committed
  55e6074

  Copy full SHA for 55e6074

• minimal protection against invalid UTF8 encoding.

  rdentato

  committed
  befe486

  Copy full SHA for befe486

• Merge pull request karpathy#267 from krrishnarraj/master

  Show description for df6557a
  karpathy

  authored
  df6557a

  Copy full SHA for df6557a

• Merge branch 'Majdoddin-ci-tiny-model'

  karpathy

  committed
  65c8993

  Copy full SHA for 65c8993

• style changes and remove spurious runc test call at the bottom

  karpathy

  committed
  62a6d69

  Copy full SHA for 62a6d69

• Merge branch 'ci-tiny-model' of https://github.com/Majdoddin/llama2.c into Majdoddin-ci-tiny-model

  karpathy

  committed
  d47fc41

  Copy full SHA for d47fc41

Commits on Aug 15, 2023

• smallfix: not sure what the point of this indirection was

  karpathy

  committed
  ca67253

  Copy full SHA for ca67253

• shorten top comment on run.c file

  karpathy

  committed
  4c63c56

  Copy full SHA for 4c63c56

• collapsing copy paste code because it's driving my ocd crazy

  karpathy

  committed
  a47f9b3

  Copy full SHA for a47f9b3

• modifiying test_all so it can safely run on windows

  Majdoddin

  committed
  87b11ed

  Copy full SHA for 87b11ed

• Adding pytest with the tiny model to macOS and windows (except amd64_arm64) runners

  Majdoddin

  committed
  66c9f5e

  Copy full SHA for 66c9f5e

• add tests into Makefile convenience

  karpathy

  committed
  88eb238

  Copy full SHA for 88eb238

• Merge pull request karpathy#297 from karpathy/feature/utf8

  Show description for 600cedb
  karpathy

  authored
  600cedb

  Copy full SHA for 600cedb

• fix sample.py from tokenizer changes before

  karpathy

  committed
  fe2de68

  Copy full SHA for fe2de68

• thoroughly commented the UTF-8 byte reading code

  karpathy

  committed
  a9a0628

  Copy full SHA for a9a0628

• add back careful processing of the byte tokens

  karpathy

  committed
  d459fd4

  Copy full SHA for d459fd4

• get rid of the special byte decoding logic

  karpathy

  committed
  4bf36ec

  Copy full SHA for 4bf36ec

• Merge branch 'utf8' of https://github.com/atamurad/llama2.c into feature/utf8

  karpathy

  committed
  8417cb4

  Copy full SHA for 8417cb4

Commits on Aug 14, 2023

• Merge branch 'master' of github.com:karpathy/llama2.c

  karpathy

  committed
  94a3a5e

  Copy full SHA for 94a3a5e

• missed p->dim to kv_dim for k,v vectors. we're not doing anything wrong we're just being wasteful with memory. thanks @xefoci7612 for pointing out

  karpathy

  committed
  32c1ff9

  Copy full SHA for 32c1ff9

• Merge pull request karpathy#286 from Nick-infinity/master

  Show description for 013e012
  karpathy

  authored
  013e012

  Copy full SHA for 013e012

• Merge pull request karpathy#289 from chenyangMl/update_readme

  Show description for 50f970d
  karpathy

  authored
  50f970d

  Copy full SHA for 50f970d

• update readme wiht a simple line to introduce llama2.c-zh

  

  chenyang

  committed
  2a9a4c4

  Copy full SHA for 2a9a4c4

• update readme wiht a simple line to introduce llama2.c-zh

  

  chenyang

  committed
  79900ff

  Copy full SHA for 79900ff

• Merge remote-tracking branch 'upstream/master'

  krrishnarraj

  committed
  eec9ad5

  Copy full SHA for eec9ad5

• remove tiktoken as dependency

  karpathy

  committed
  82ad2ba

  Copy full SHA for 82ad2ba

• [Feat]: Add support for meta llama hf model conversion

  Show description for c39f19f
  Nick-infinity

  committed
  c39f19f

  Copy full SHA for c39f19f

• Small tweaks to Readme intro

  karpathy

  authored
  bae0bcf

  Copy full SHA for bae0bcf

• the accum function has been bothering me, there is no real need to add a function here, it does something trivial and is only used twice, scrap

  karpathy

  committed
  45afa91

  Copy full SHA for 45afa91

• turn topp 0.9 back on by default thanks to recent PR contributions truncating before quicksort

  karpathy

  committed
  854c97b

  Copy full SHA for 854c97b

• Merge pull request karpathy#276 from jrudolph/improve-top-p

  Show description for 4a2c375
  karpathy

  authored
  4a2c375

  Copy full SHA for 4a2c375

Pagination

Previous

Next