 oobabooga
|
c0f2367b54
Minor fix
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
185587a33e
Add a history size parameter to the chat
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
78d5a999e6
Improve prompt formatation
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
70ff685736
Encode the input string correctly
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
b66d18d5a0
Allow presets/characters with '.' in their names
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
11c3214981
Fix some regexes
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
e61138bdad
Minor fixes
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
2181fca709
Better defaults for chat
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
83808171d3
Add --share option for Colab
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
8d788874d7
Add support for characters
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
3121f4788e
Fix uploading chat log in --chat mode
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
849e4c7f90
Better way of finding the generated reply in the output string
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
d03b0ad7a8
Implement saving/loading chat logs (#9)
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
39bfea5a22
Add a progress bar
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
5390fc87c8
add auto-devices when disk is used
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
759da435e3
Release 8-bit models memory
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
7ace04864a
Implement sending layers to disk with --disk (#10)
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
93fa9bbe01
Clean up the streaming implementation
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
c90310e40e
Small simplification
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
99536ef5bf
Add no-stream option
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
116299b3ad
Manual eos_token implementation
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
3cb30bed0a
Add a "stop" button
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
8f27d33034
Fix another bug
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
6c7f187586
Minor change
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
b3cba0b330
Bug
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
df2e910421
Stop generating in chat mode when \nYou: is generated
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
022960a087
This is the correct way of sampling 1 token at a time
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
0f01a3b1fa
Implement text streaming (#10)
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
ca13acdfa0
Ensure that the chat prompt will always contain < 2048 tokens
|
%!s(int64=3) %!d(string=hai) anos |
|
oobabooga
|
6456777b09
Clean things up
|
%!s(int64=3) %!d(string=hai) anos |
