Commit History
keep gate in fp32 for 16 bit loras (#1105)
da97285
unverified
winglian
commited on
feat: enable trl's autounwrap (#1060)
b432889
unverified
Nanobit
commited on
fix model card upload for PEFT models (#1043)
31d2350
unverified
hamel
commited on
RL/DPO (#935)
f243c21
winglian
commited on
add config to model card (#1005)
85dd4d5
unverified
hamel
commited on
fix: switch to using the HuggingFace Transformers NEFT implementation (#941)
ef24342
unverified
dg-kalle
commited on
Fix Deepspeed loading (#950)
5ea3aa3
unverified
winglian
commited on
support for mamba (#915)
40a6362
unverified
winglian
commited on
use accelerate logging for zero/main loggin only
b2430ce
winglian
commited on
cleanup verbosity a bit
4c834bf
winglian
commited on
refactor neft patch to be more re-usable similar to trl's impl (#796)
827ec3d
unverified
winglian
commited on
Implement fused modules (#747)
15d3a65
unverified
Fix DeepSpeed Zero 3 Saving (#709)
e4d1585
unverified
create a model card with axolotl badge (#624)
501958b
unverified
winglian
commited on
set fsdp state dict (#584)
be75668
unverified
Jan Philipp Harries
Jan Philipp Harries
commited on
misc fixes/improvements (#513)
a546ca2
unverified
winglian
commited on
split train from other cli options (#503)
b21e4a2
unverified
winglian
commited on