auto-round
fix lots of gguf critic bugs and support imatrix in rtn mode
#595
Merged
Merged
fix lots of gguf critic bugs and support imatrix in rtn mode #595
wenhuach21
merged
44 commits into
main
from
imatrix
try to setup imatrix baseline
4c79e5e6
add todo
7731c2d6
![pre-commit-ci[bot]](https://avatars.githubusercontent.com/in/68672?v=4&s=60)
[pre-commit.ci] auto fixes from pre-commit.com hooks
7eaf8e41
update
f8533b2f
Merge branch 'main' into imatrix
277a2d94
Merge branch 'imatrix' of https://github.com/intel/auto-round into im…
49b381e3
tmp change
b02895fd
update
c428e2b0
[pre-commit.ci] auto fixes from pre-commit.com hooks
a92b36ac
tiny change
2650b41e
clear memory
cd0f01ff
avg the act
1f9bbe75
quantize embedding and lm_head in rtn mode
ae7cda1d
wenhuach21
changed the title
fix bugs
1caf541e
support 3bits and 6 bits imatrix
9076241e
fix multi gpu issue
10af3b9a
Tried to fix the MoE issue, but it hasn't been addressed yet.
b4b83d09
fix
90863c87
tmp change
3088133f
tmp change
8313d2c4
fix moe issue
095f113a
wenhuach21
changed the title
fix q2_k_s tiny
28dae2dd
wenhuach21
changed the title
fix and support embedding quantization in tuning
36718b28
support lm-head tuning for gguf by default
48c4adad
[pre-commit.ci] auto fixes from pre-commit.com hooks
d867161c
fix permute issue
dc58cb64
unify packing code for different devices
c4a13e52
fix issues
3c3dbd39
fix issues
a9bc4eb7
fix issues
a76fdc00
wenhuach21
changed the title 
copilot-pull-request-reviewer
commented
on 2025-06-13
fix
ffa23df8
fix
7b544987
fix
dd070ef7
fix
844c0f18
fix one ut
706f2ea0
WeiweiZhang1
approved these changes
on 2025-06-16
n1ck-guo
approved these changes
on 2025-06-16
support fallback and fix ut
1980bc29
fix ut and add disable_opt_rtn
50c933ba
change one file name
8a54559c
porting mixed bits setting from gguf
9311915b
fix
2099091c
fix
f096c199
fix
56fe2874
fix
5651b798
fix
425af268
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Login to write a write a comment.
Login via GitHub