Add OLMoE
082973c1
Add OLMoE
85885764
Updates
f1c569e6
Make norm optional; add keys
f6ea7c51
Add output
4d567220
Add
8b176d9b
Fix dtype
452da8d9
Fix eos config
140bafb1
Update
91f95fde
Add OLMoE
6c20b738
git pushMerge branch 'olmoe' of https://github.com/Muennighoff/transf…
171602ec
Fix OLMoE path
30a4feb8
Merge branch 'huggingface:main' into olmoe
698f1566
Format
474f8e82
git stah popMerge branch 'olmoe' of https://github.com/Muennighoff/tr…
e7e2ce36
Format
d3eeef00
Rmv copy statement
28cdfd8c
Rmv copy statement
58aed4a4
Format
f9fbd12b
Add copies
16ed9e1a
Cp rotary
b9a045aa
Fix aming
4c598be6
Fix naming
50507eac
Merge branch 'huggingface:main' into olmoe
1d9b006b
Update RoPE integration; num_logits_to_keep; Add copy statements
b9948cc8
Add eps to config
e97ae0e1
Format
fd0baf50
Add aux loss
79e0ecc4
Adapt router_aux_loss_coef
758a808b
Update md
efdcda6f
Merge branch 'huggingface:main' into olmoe
42145af0
Adapt
34ef8f55
adapt tests
30aace4b
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub