Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
jinaai
/
jina-bert-flash-implementation
like
1
Transformers
bert
custom_code
Inference Endpoints
๐ช๐บ Region: EU
Model card
Files
Files and versions
Community
18
Train
Deploy
Use this model
refs/pr/16
jina-bert-flash-implementation
6 contributors
History:
108 commits
Markus28
Set activation_checkpoint_lvl to 100 by default
535ad9a
4 months ago
README.md
1.89 kB
feat: added README
4 months ago
bert_padding.py
9.78 kB
reference the flash attention GitHub
5 months ago
block.py
17.4 kB
reference the flash attention GitHub
5 months ago
configuration_bert.py
7.33 kB
Set activation_checkpoint_lvl to 100 by default
4 months ago
convert_v2_weights.py
6.1 kB
feat: for converting v2, added lines to save model weights and print config
4 months ago
embedding.py
2.26 kB
clean up embeddings.py (#7)
5 months ago
mha.py
35.3 kB
reference the flash attention GitHub
5 months ago
mlp.py
7.79 kB
Porting v2 models to flash attention (#15)
4 months ago
modeling_bert.py
33.9 kB
feat: added selective activation checkpointing
4 months ago
modeling_for_glue.py
10.7 kB
feat: assert return_dict
5 months ago
modeling_lora.py
12.3 kB
fix: use staticmethod istead of classmethod
4 months ago