PhiMoE #33363
ArthurZucker
merged
31 commits into
huggingface:main
from
garg-amit:gargamit/onboard_phi3_5_moe
onboard phimoe model
6ad863ab
removed debug code
1a1e547f
added unit tests
08d73d78
updated docs
1277bc8b
formatted
232588d3
fixed unit tests
84627838
fixed test case
3668c5d2
fixed format
e6ed8dc4
refactored code
89f51eab
fixed expected outputs in the integration tests
e552e335
Added a warning msg
c8173d75
Merge branch 'main' into gargamit/onboard_phi3_5_moe
290514e5
Merge branch 'main' into gargamit/onboard_phi3_5_moe
5dda7d64
Merge branch 'gargamit/onboard_phi3_5_moe' of https://github.com/garg…
11a0f176
Merge branch 'main' of https://github.com/garg-amit/transformers into…
dd02bf9d
Addressed comments
43f9cc94
Merge branch 'main' of https://github.com/garg-amit/transformers into…
b3f8af5b
Addressed comments
b6acc3e0
fixed test cases
33caa631
added paper link
e01a78e4
Addressed comments
d1f847ef
Refactored PhimoeForCausalLM forward fn
dd8b8b01
Refactored PhimoeRotaryEmbedding class
42b59c69
fixed test cases
e4e2f1a5
fixed testcase
8359a598
fixed test case
bebad97f
Merge branch 'main' of https://github.com/garg-amit/transformers into…
4c28bd55
Addressed comments
1311e80a
fixed test cases
2671887c
fixed testcases
18830f59
ArthurZucker
approved these changes
on 2024-10-04
Used cache position instead to get the seq len
36891bfc
Login to write a write a comment.
Login via GitHub