norbert4-large / modeling_gptbert.py

Commit History

make FlashAttention logic more robust
6f3b747
verified

davda54 commited on

fix
fcc9a7d
verified

davda54 commited on

removed SDPA
d30e2d1
verified

davda54 commited on

Update modeling_gptbert.py
232b117
verified

davda54 commited on

fixed SDPA for older PyTorch versions
85a2222
verified

davda54 commited on

fix
e419f0f
verified

davda54 commited on

Update modeling_gptbert.py
8b27574
verified

davda54 commited on

Update modeling_gptbert.py
042ceea
verified

davda54 commited on

fix
30fd0e7
verified

davda54 commited on

fix
39265fc
verified

davda54 commited on

fix
87e0acb
verified

davda54 commited on

fix attention mask
16b2b5e
verified

davda54 commited on

fix
6cbbc37
verified

davda54 commited on

fix
37fc2ed
verified

davda54 commited on

Update modeling_gptbert.py
60e56bf
verified

davda54 commited on

Update modeling_gptbert.py
7741265
verified

davda54 commited on

fixes and optimizations
3bd5fb5
verified

davda54 commited on

Fix import error
e071182
verified

davda54 commited on

Update modeling_gptbert.py
268e841
verified

lgcharpe commited on

Update modeling_gptbert.py
2d600a4
verified

lgcharpe commited on

Update modeling_gptbert.py
9e04fd3
verified

lgcharpe commited on

Update modeling_gptbert.py
45e963e
verified

lgcharpe commited on

Update modeling_gptbert.py
91e4060
verified

lgcharpe commited on

Update modeling_gptbert.py
a36b0d0
verified

lgcharpe commited on

Update modeling_gptbert.py
4abdbe5
verified

lgcharpe commited on

Update modeling_gptbert.py
f5214f6
verified

lgcharpe commited on

Checking for errors in the CastedLinear
3e3d609
verified

lgcharpe commited on

Update modeling_gptbert.py
73e5534
verified

lgcharpe commited on

More testing
74cf691
verified

lgcharpe commited on

Add checking for NaN
64bbd9e
verified

lgcharpe commited on

Removing added print
757ab94
verified

lgcharpe commited on

Testing output
98c5a3b
verified

lgcharpe commited on

fix flasshattention
64f341a
verified

davda54 commited on

test
c96534d
verified

davda54 commited on

test
6a98774
verified

davda54 commited on

fix attention masks
e228931
verified

davda54 commited on

fix import error
0fd4c6d
verified

davda54 commited on

Fix error
7ee002c
verified

davda54 commited on

Add FlashAttention + unpadding support
4d4d26d
verified

davda54 commited on

Update modeling_gptbert.py
28e46d7
verified

lgcharpe commited on

updated MLM predictions
790a5cc
verified

davda54 commited on

Pass throught _init_weights
45f6842
verified

davda54 commited on

Fixed classifier head
4e70337
verified

davda54 commited on

Upload folder using huggingface_hub
a3ddaaf
verified

davda54 commited on