Commits · jinaai/jina-bert-flash-implementation

feat: added cleaved_layers property

4441ce6

Markus28 commited on Mar 15

feat: added functionality to cleave off layers from BERT encoder

86b0438

Markus28 commited on Mar 15

fix BertForMaskedLM

c0b46cc

Markus28 commited on Mar 14

feat: added separate BertForMaskedLM class

3cb3930

Markus28 commited on Mar 14

Update tokenizer.py

9072f7f
verified

michael-guenther commited on Mar 13

feat-add-constant-for-task-type-ids (#10)

11b09c9
verified

michael-guenther commited on Mar 13

feat: added return_dict

59c0808

Markus28 commited on Mar 12

fix: fixed syntax error in LoRA

e93b0fd

Markus28 commited on Mar 12

feat: add current_task to forward

9410275

Markus28 commited on Mar 12

feat: use property in LoRA parametrization

0ff7c3d

Markus28 commited on Mar 11

feat: added LoRA copyright notice

faa9951

Markus28 commited on Mar 11

feat: use property instead of setter

6aad619

Markus28 commited on Mar 11

clean up embeddings.py (#7)

7771ce3
verified

Markus28

bwang0911 commited on Mar 11

feat: return from_bert for from_pretrained

5549314

Markus28 commited on Mar 7

support-fast-tokenizer (#6)

ed1b276
verified

jupyterjazz commited on Mar 7

feat: made from_bert work

851184a

Markus28 commited on Mar 6

feat: choose flash attention heuristically if not set explicitly

2e2b8d0

Markus28 commited on Mar 6

feat: select first LoRA upon initialization

fabeb13

Markus28 commited on Mar 6

feat: formatting and type hints

617fe56

Markus28 commited on Mar 6

support-multiple-task-ids (#5)

e151a8f
verified

Markus28

michael-guenther commited on Mar 6

fix: use proper initilization for embedding layer

850b9a2

Markus28 commited on Mar 6

fix: fixed typo

5c4e4bf

Markus28 commited on Mar 6

feat: added LoRA

8561a1f

Markus28 commited on Mar 6

feat: assert return_dict

326b1c4

Markus28 commited on Mar 5

fix: same assertions in other models

c1d92c9

Markus28 commited on Mar 5

fix: assert is None for other kwargs too

3f5615c

Markus28 commited on Mar 5

feat: added head_mask

599c64e

Markus28 commited on Mar 5

added classifier dropout

767b681

Markus28 commited on Mar 5

fix: formatting

ae4c28c

Markus28 commited on Mar 5

fix: formatting

f115a1d

Markus28 commited on Mar 5

feat: added further GLUE models

ec37ae5

Markus28 commited on Mar 5

feat: added BertForSequenceClassification

ba24fb1

Markus28 commited on Mar 5

fix: cast mask to bool

ca5f516

Markus28 commited on Mar 5

reference the flash attention GitHub

eec6c0e

Markus28 commited on Mar 5

fix: move flash components into top-level

5944ec8

Markus28 commited on Mar 5

feat: try to fix import error

4c4562b

Markus28 commited on Mar 5

feat: moved flash attention code into this repository

46df05d

Markus28 commited on Mar 5

add tokenizer

6343db7

michael-guenther commited on Mar 4

feat: added encode method

32458be

Markus28 commited on Mar 1

fix: try to skip initialization of task type embeddings

3b35eab

Markus28 commited on Mar 1

fix: try to skip initialization of task type embeddings

95ca1a8

Markus28 commited on Mar 1

feat: added option for QK normalization

463061d

Markus28 commited on Mar 1

fix: removed obscure config options

2e69073

Markus28 commited on Mar 1

feat: added small config

149d26f

Markus28 commited on Mar 1

feat: implement task type embeddings (#1)

8adf551
verified

Markus28 commited on Mar 1

feat: added back option not to use flash attention

d4d5621

Markus28 commited on Mar 1

feat: support gradient checkpointing

75d7a16

Markus28 commited on Feb 28

Added additional config options

5b58f09

Markus28 commited on Feb 27

removed unused imports

5e7b835

Markus28 commited on Feb 27

removed init from BertPretrainedModel

44fd417

Markus28 commited on Feb 27

Commit History

feat: added cleaved_layers property 4441ce6

feat: added functionality to cleave off layers from BERT encoder 86b0438

fix BertForMaskedLM c0b46cc

feat: added separate BertForMaskedLM class 3cb3930

Update tokenizer.py 9072f7f verified

feat-add-constant-for-task-type-ids (#10) 11b09c9 verified

feat: added return_dict 59c0808

fix: fixed syntax error in LoRA e93b0fd

feat: add current_task to forward 9410275

feat: use property in LoRA parametrization 0ff7c3d

feat: added LoRA copyright notice faa9951

feat: use property instead of setter 6aad619

clean up embeddings.py (#7) 7771ce3 verified

feat: return from_bert for from_pretrained 5549314

support-fast-tokenizer (#6) ed1b276 verified

feat: made from_bert work 851184a

feat: choose flash attention heuristically if not set explicitly 2e2b8d0

feat: select first LoRA upon initialization fabeb13

feat: formatting and type hints 617fe56

support-multiple-task-ids (#5) e151a8f verified

fix: use proper initilization for embedding layer 850b9a2

fix: fixed typo 5c4e4bf

feat: added LoRA 8561a1f

feat: assert return_dict 326b1c4

fix: same assertions in other models c1d92c9

fix: assert is None for other kwargs too 3f5615c

feat: added head_mask 599c64e

added classifier dropout 767b681

fix: formatting ae4c28c

fix: formatting f115a1d

feat: added further GLUE models ec37ae5

feat: added BertForSequenceClassification ba24fb1

fix: cast mask to bool ca5f516

reference the flash attention GitHub eec6c0e

fix: move flash components into top-level 5944ec8

feat: try to fix import error 4c4562b

feat: moved flash attention code into this repository 46df05d

add tokenizer 6343db7

feat: added encode method 32458be

fix: try to skip initialization of task type embeddings 3b35eab

fix: try to skip initialization of task type embeddings 95ca1a8

feat: added option for QK normalization 463061d

fix: removed obscure config options 2e69073

feat: added small config 149d26f

feat: implement task type embeddings (#1) 8adf551 verified

feat: added back option not to use flash attention d4d5621

feat: support gradient checkpointing 75d7a16

Added additional config options 5b58f09

removed unused imports 5e7b835

removed __init__ from BertPretrainedModel 44fd417

feat: added cleaved_layers property

4441ce6

feat: added functionality to cleave off layers from BERT encoder

86b0438

fix BertForMaskedLM

c0b46cc

feat: added separate BertForMaskedLM class

3cb3930

Update tokenizer.py

9072f7f
verified

feat-add-constant-for-task-type-ids (#10)

11b09c9
verified

feat: added return_dict

59c0808

fix: fixed syntax error in LoRA

e93b0fd

feat: add current_task to forward

9410275

feat: use property in LoRA parametrization

0ff7c3d

feat: added LoRA copyright notice

faa9951

feat: use property instead of setter

6aad619

clean up embeddings.py (#7)

7771ce3
verified

feat: return from_bert for from_pretrained

5549314

support-fast-tokenizer (#6)

ed1b276
verified

feat: made from_bert work

851184a

feat: choose flash attention heuristically if not set explicitly

2e2b8d0

feat: select first LoRA upon initialization

fabeb13

feat: formatting and type hints

617fe56

support-multiple-task-ids (#5)

e151a8f
verified

fix: use proper initilization for embedding layer

850b9a2

fix: fixed typo

5c4e4bf

feat: added LoRA

8561a1f

feat: assert return_dict

326b1c4

fix: same assertions in other models

c1d92c9

fix: assert is None for other kwargs too

3f5615c

feat: added head_mask

599c64e

added classifier dropout

767b681

fix: formatting

ae4c28c

fix: formatting

f115a1d

feat: added further GLUE models

ec37ae5

feat: added BertForSequenceClassification

ba24fb1

fix: cast mask to bool

ca5f516

reference the flash attention GitHub

eec6c0e

fix: move flash components into top-level

5944ec8

feat: try to fix import error

4c4562b

feat: moved flash attention code into this repository

46df05d

add tokenizer

6343db7

feat: added encode method

32458be

fix: try to skip initialization of task type embeddings

3b35eab

fix: try to skip initialization of task type embeddings

95ca1a8

feat: added option for QK normalization

463061d

fix: removed obscure config options

2e69073

feat: added small config

149d26f

feat: implement task type embeddings (#1)

8adf551
verified

feat: added back option not to use flash attention

d4d5621

feat: support gradient checkpointing

75d7a16

Added additional config options

5b58f09

removed unused imports

5e7b835

removed init from BertPretrainedModel

44fd417