jinaai
/

jina-bert-flash-implementation

🇪🇺 Region: EU

Model card Files Files and versions Community

jina-bert-flash-implementation

6 contributors

History: 106 commits

Markus28's picture

fixed GLU implementation, added conversion of layer norms

9587227 about 1 year ago

README.md

1.89 kB

feat: added README about 1 year ago
bert_padding.py

9.78 kB

reference the flash attention GitHub about 1 year ago
block.py

17.4 kB

reference the flash attention GitHub about 1 year ago
configuration_bert.py

5.77 kB

Added GLUMLP, changed config accordingly, added code to convert state_dict about 1 year ago
convert_v2_weights.py

6.01 kB

fixed GLU implementation, added conversion of layer norms about 1 year ago
embedding.py

2.26 kB

clean up embeddings.py (#7) about 1 year ago
mha.py

35.3 kB

reference the flash attention GitHub about 1 year ago
mlp.py

7.79 kB

fixed GLU implementation, added conversion of layer norms about 1 year ago
modeling_bert.py

33.4 kB

Added GLUMLP, changed config accordingly, added code to convert state_dict about 1 year ago
modeling_for_glue.py

10.7 kB

feat: assert return_dict about 1 year ago
modeling_lora.py

12.3 kB

fix: use staticmethod istead of classmethod about 1 year ago