Vulcan
|
45e0fa941f
|
Update CITATION.cff
|
2023-10-02 19:17:24 +05:30 |
|
Vulcan
|
ae8ea66eda
|
L2E OS v0.1 "Temple DOS" Release + Various Improvements
L2E OS v0.1 "Temple DOS" Release.
Various Improvements.
Major Makefile overhaul.
|
2023-10-02 19:03:18 +05:30 |
|
Adarsh Shirawalmath
|
9fdb1316c7
|
Update README.md
|
2023-10-01 11:18:33 +05:30 |
|
Nicky Pochinkov
|
2dedad6cea
|
Added support for repeated kv weights
|
2023-09-21 16:38:06 +02:00 |
|
Nicky Pochinkov
|
d3c25b10a6
|
Add checks/config for tied embedding weights
|
2023-09-21 16:36:36 +02:00 |
|
Nicky Pochinkov
|
ffea287516
|
updated comment .pt -> .bin
|
2023-09-16 18:46:27 +01:00 |
|
Nicky Pochinkov
|
a61173d6b9
|
Added CLI dtype code
|
2023-09-16 18:32:31 +01:00 |
|
Nicky Pochinkov
|
19f40a2a71
|
Made default hf export torch.float32
|
2023-09-16 18:32:21 +01:00 |
|
Nicky Pochinkov
|
fc11cc387b
|
Changed code so that lm_head and token_embed are tied
|
2023-09-16 18:10:36 +01:00 |
|
Nicky Pochinkov
|
f38055dfb6
|
add option to set dtype for export
|
2023-09-16 14:07:48 +01:00 |
|
jackpeck
|
b259fb4432
|
Add link to pure OCaml port
|
2023-09-16 13:43:10 +01:00 |
|
Bernardo Ramos
|
411c5bd2db
|
reorganize variables
|
2023-09-14 07:14:45 +00:00 |
|
Bernardo Ramos
|
593d846bc3
|
use key and value from kv cache
|
2023-09-14 01:13:08 +00:00 |
|
Bernardo Ramos
|
3da6cc1b21
|
readme: add another javascript port
|
2023-09-13 16:09:39 -03:00 |
|
Nicky Pochinkov
|
bf9a1162e1
|
Added error handling for LlamaConfig import
|
2023-09-12 19:55:28 +01:00 |
|
Nicky Pochinkov
|
6360a53901
|
fixed whitespace
|
2023-09-12 19:53:26 +01:00 |
|
Nicky Pochinkov
|
c568f6952d
|
added option to export to huggingface format
|
2023-09-12 19:51:31 +01:00 |
|
Andrej
|
766a30bc6e
|
Merge pull request #391 from karpathy/revert-389-realtime
Revert "Minor fix: Use CLOCK_MONOTONIC instead of CLOCK_REALTIME"
|
2023-09-12 09:18:43 +01:00 |
|
Andrej
|
38c58ac336
|
Revert "Minor fix: Use CLOCK_MONOTONIC instead of CLOCK_REALTIME"
|
2023-09-12 11:17:04 +03:00 |
|
Andrej
|
0bf2981793
|
Merge pull request #387 from flaneur2020/add-notable-forks
chore: add another rust fork to the notable forks section
|
2023-09-12 09:13:28 +01:00 |
|
Andrej
|
27e623c8ae
|
Merge pull request #388 from tairov/patch
Add link to pure Mojo implementation of project
|
2023-09-12 09:12:31 +01:00 |
|
Andrej
|
d15577b541
|
Merge pull request #389 from jbochi/realtime
Minor fix: Use CLOCK_MONOTONIC instead of CLOCK_REALTIME
|
2023-09-12 09:11:58 +01:00 |
|
Juarez Bochi
|
9d73a377fb
|
Use CLOCK_MONOTONIC instead of realtime
|
2023-09-11 14:05:37 -04:00 |
|
Aydyn Tairov
|
38011d070a
|
Add link to pure Mojo port of project
|
2023-09-11 12:29:25 +01:00 |
|
Li Yazhou
|
3ac620572e
|
add another rust implementation
|
2023-09-10 21:22:05 +08:00 |
|
Andrej
|
35deb5e0fa
|
Merge pull request #370 from dmarcos/llama2c-web
Add llama2.c-web to the list of projects in Readme.md
|
2023-09-06 22:55:53 +01:00 |
|
Andrej
|
5186b505f3
|
Merge pull request #383 from atamurad/int8_refactor
int8 refactor
|
2023-09-05 23:36:12 +01:00 |
|
Andrej
|
3b0a6d3927
|
Merge pull request #380 from dfurrer/master
Remove duplicate word in comment. EOM
|
2023-09-05 06:54:18 -04:00 |
|
Andrej
|
358fe99150
|
Merge pull request #381 from rbitr/edit_clones
added fortran clone
|
2023-09-05 06:54:06 -04:00 |
|
Andrew
|
0b3a5e17fd
|
added fortran clone
|
2023-09-03 17:54:54 -04:00 |
|
Daniel Furrer
|
a69ee269c5
|
Update run.c
Remove duplicate word in comments.
|
2023-09-03 22:37:10 +02:00 |
|
Vulcan
|
733a625b29
|
Update README.md
|
2023-09-02 10:04:40 +05:30 |
|
Vulcan
|
7e589f83a6
|
Merge remote-tracking branch 'upstream/master'
|
2023-09-02 09:22:48 +05:30 |
|
Andrej
|
b9fb86169f
|
Merge pull request #367 from janimo/long-multiply
Do parameter count calculations in 64 bits to not overflow in case of…
|
2023-09-01 10:07:12 -07:00 |
|
Andrej
|
0776f86793
|
Merge pull request #374 from SpaceCowboy850/bugfix_train_vocab
Setting an encoding for tiny_file tokenizer file.
|
2023-09-01 09:51:50 -07:00 |
|
Brandon Rowlett
|
ab19aa0804
|
Setting UTF encoding, otherwise windows breaks with UnicodeEncodeError: 'charmap' codec can't encode character '\u200b' in position 971: character maps to <undefined>
|
2023-08-30 14:54:41 -05:00 |
|
Jani Monoses
|
c5ec6e21b8
|
Use long long so it works with MSVC
|
2023-08-29 17:47:55 +03:00 |
|
Diego Marcos Segura
|
fab753db3a
|
Add llama2.c-web to the list of projects in README.md
|
2023-08-28 14:48:38 -07:00 |
|
Vulcan
|
9b6beee952
|
Create FUNDING.yml
|
2023-08-28 15:35:48 +05:30 |
|
Vulcan
|
2d9697b34d
|
win64 build fix
LSTATUS changed to LOOPSTATUS.
LSTATUS must have been defined somewhere in windows related headers.
|
2023-08-28 13:42:39 +05:30 |
|
Vulcan
|
c3210fa0ed
|
win64 & unikernel build fixes
|
2023-08-28 13:31:54 +05:30 |
|
Vulcan
|
eb9d77f0d3
|
Merge remote-tracking branch 'upstream/master'
|
2023-08-28 11:53:37 +05:30 |
|
Jani Monoses
|
1ebb27f090
|
Do parameter count calculations in 64 bits to not overflow in case of very large models
|
2023-08-27 12:21:11 +03:00 |
|
atamyrat
|
6e52df9b41
|
properly handle token embeddings & shared classifier wcls
|
2023-08-27 08:18:03 +03:00 |
|
atamyrat
|
06175b946b
|
free() quantizedtensors
|
2023-08-27 06:47:03 +03:00 |
|
Andrej
|
7325bab657
|
Merge pull request #365 from atamurad/patch-1
Update README.md - unclosed code block quotes
|
2023-08-26 20:11:04 -07:00 |
|
atamyrat
|
f850a97c6a
|
draft refactor to use QuantizedTensor in function arguments
|
2023-08-27 06:05:20 +03:00 |
|
Atamurad Hezretkuliyev
|
37157bc0a3
|
Update README.md
Fixed unclosed code block quotes
|
2023-08-27 02:27:47 +03:00 |
|
Andrej Karpathy
|
df80471914
|
draft of int8 attempt number two
|
2023-08-26 22:28:08 +00:00 |
|
Andrej Karpathy
|
f4b8a81742
|
Merge branch 'master' of github.com:karpathy/llama2.c
|
2023-08-26 21:22:28 +00:00 |
|