Commit History

0.24 Load in 8bit
65118fd

5to9 commited on

0.24 Log to nvidia-smi
0e52052

5to9 commited on

0.24 Set device map to auto. remove eval
c5570f6

5to9 commited on

0.23 Set device map to cuda
0c375c9

5to9 commited on

0.22 remove subprocess flash attn. Set device map balanced
2c2a26f

5to9 commited on

0.21 remove subprocess flash attn. Force device map
128c5e3

5to9 commited on

0.20 subprocess implementing flash_attn
1c109f3

5to9 commited on

0.19 implementing flash_attn
20d9962

5to9 commited on

0.18 bfloat to float
1a0b767

5to9 commited on

0.17 changed requirements
7444257

5to9 commited on

0.14 explictly set GPU and dtype
b235bfd

5to9 commited on

0.14 set dtype of input_ids
d0aacc5

5to9 commited on

0.13 catch more exceptions
ee6ec78

5to9 commited on

0.12 catch exceptions
b6c4ccb

5to9 commited on

0.11 simplifying wo pharia
75b1a69

5to9 commited on

0.10 removing flash attn.
2569b24

5to9 commited on

0.9 logging template func
eea02ee

5to9 commited on

0.8 fixed wrong device
19ef1ca

5to9 commited on

0.7 fixing case typo pharia to Pharia
f0039d0

5to9 commited on

0.6 defining chat template for pharia
c6e8ef5

5to9 commited on

0.5 tweaking output, ltr
7977c5d

5to9 commited on

0.4 adding new example prompt ltr
992ab1c

5to9 commited on

0.3 resolve dependcy fckp
7d7914e

5to9 commited on

0.2 adding cloned space
74ffc36

5to9 commited on

initial commit
92c6271
verified

5to9 commited on

Duplicate from gradio-templates/chatbot
dbf2aa5
verified

5to9 pngwn HF staff commited on