 Matthias Reso
|
c5a382e509
Make tests run on cpu only machines
|
%!s(int64=2) %!d(string=hai) anos |
 Hamid Shojanazeri
|
98fcc538ff
Add option to enable Llamaguard content safety check in chat_completion (#354)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
d89a02feea
Fix Anyscale API token URL in Purple_Llama_Anyscale.ipynb (#362)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
0af79e0263
Remove deprecated pytest_cmdline_preparse (#363)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
d9d04a1a66
Add script to run cloud api throughput benchmark (#339)
|
%!s(int64=2) %!d(string=hai) anos |
 Chester Hu
|
6d44371c99
Update README.md
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
e9985fc4ad
adding HSDP as sharding strategy for FSDP training composable with PEFT (#281)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
43ea6bfa71
Merge branch 'main' into ssdp
|
%!s(int64=2) %!d(string=hai) anos |
|
Chester Hu
|
986847f685
Merge branch 'main' into benchmark-inference-throughput-cloud-api
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
622200340e
Codellama 70b instruct code example (#365)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
d51d2cce9c
adding sdpa for flash attn
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
db8af96ff0
update the model load with native flash attn
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
9a2434f408
update the fine-tuning script
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
8776ceb833
update with the HF flash attention native
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
6853267317
update with the HF flash attention native
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
0854a74df1
update readme
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
ce966a97e0
add native BT
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
59b8e3a5a3
update main readme
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
3484a06b01
update main readme
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
9d5fd77d70
fix typos
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
273baf58f0
update READMEs
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
78315e9eb0
adding llama70b instruct code example
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
a2f5b66b1f
update the code with llama-guard safety check
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
85ee7bcc24
update the code with llama-guard safety check
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
dae93e3da7
update inference readme for codellama 70B instruct
|
%!s(int64=2) %!d(string=hai) anos |
|
Matthias Reso
|
147aaa29bc
Remove deprecated pytest_cmdline_preparse
|
%!s(int64=2) %!d(string=hai) anos |
 dramdass
|
f0fdd3c4fa
Fix Anyscale API token URL in Purple_Llama_Anyscale.ipynb
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
ad5ff2dcb8
Revert "Flop counter, profiling and GC (#357)" (#361)
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
162be4c045
Revert "Flop counter, profiling and GC (#357)"
|
%!s(int64=2) %!d(string=hai) anos |
|
Hamid Shojanazeri
|
2145e4729f
Eval harness (#356)
|
%!s(int64=2) %!d(string=hai) anos |
