changes done according to llama model
173b5f1
Lahiru Menikdiwela
commited on
Commit History
add max new tokens to pipeline
ab21bba
Lahiru Menikdiwela
commited on
fix error related to model kwargs
652e925
Lahiru Menikdiwela
commited on
add model kwargs for llama
f40f333
Lahiru Menikdiwela
commited on
remove quantization due to lack of cuda
3ef330a
Lahiru Menikdiwela
commited on
load model in 4bit and bfloat16 computying type with pipeline op check
638094e
Lahiru Menikdiwela
commited on
fix automodel for llama - text generation
9a32a9c
Lahiru Menikdiwela
commited on
fix automodel for llama -pipepline
70f6a45
Lahiru Menikdiwela
commited on
fix automodel for llama
72b02de
Lahiru Menikdiwela
commited on
revert changes
16ecc46
Lahiru Menikdiwela
commited on
trying to fix config.json issues in led
32e6acc
Lahiru Menikdiwela
commited on
