Commits · jinaai/xlm-roberta-flash-implementation

fix: update frequencies when updating the rope base value (#40)

8b2ad1e
verified

jupyterjazz commited on Sep 3, 2024

feat: update the readme

a6bb16f

nan commited on Aug 30, 2024

fix mixed precision loading with recent transformers versions (#39)

bda5e8d
verified

jupyterjazz commited on Aug 27, 2024

lora-instructions (#36)

a709b51
verified

michael-guenther

Jackmin108 commited on Aug 27, 2024

fix-device-issues (#38)

57dbe22
verified

jupyterjazz

Jackmin108 commited on Aug 26, 2024

fix: override use_flash_attn in lora

dc4080e
verified

jupyterjazz commited on Aug 26, 2024

Fix rotary backward

169b7fb
verified

jupyterjazz commited on Aug 23, 2024

cpu-inference (#35)

80b1ff7
verified

jupyterjazz

Jackmin108 commited on Aug 23, 2024

feat: configurable use_reentrant (#37)

09dbf45
verified

jupyterjazz

gmastrapas commited on Aug 22, 2024

refine-codebase (#33)

2646361
verified

jupyterjazz commited on Aug 16, 2024

fix-adapter-masks (#32)

0f0bed6
verified

jupyterjazz

Jackmin108 commited on Aug 12, 2024

change rotary base (#31)

4434bf3
verified

jupyterjazz

Jackmin108 commited on Aug 12, 2024

2-adapter-tuning (#29)

95fd08c
verified

jupyterjazz

Jackmin108 commited on Aug 12, 2024

Update modeling_xlm_roberta.py

027219a
verified

jupyterjazz commited on Aug 7, 2024

2-adapter-tuning-initial-impl (#30)

2cc8472
verified

jupyterjazz

Jackmin108 commited on Aug 7, 2024

Update modeling_xlm_roberta.py

3eceb33
verified

jupyterjazz commited on Aug 6, 2024

make-lora-stateless (#25)

e860caa
verified

jupyterjazz commited on Aug 5, 2024

feat-routing (#26)

6a92924
verified

jupyterjazz

makram93 commited on Aug 2, 2024

fix-rope-inference (#28)

955fea2
verified

jupyterjazz commited on Aug 1, 2024

no-flash-attention-during-inference (#22)

7ad815b
verified

jupyterjazz commited on Jul 23, 2024

Rename config to config.json

4d09ca8
verified

jupyterjazz commited on Jul 16, 2024

Create config

721e106
verified

bwang0911 commited on Jul 1, 2024

Delete config.json

a60f7c0
verified

bwang0911 commited on Jun 30, 2024

rope-embeddings (#20)

e3681c2
verified

jupyterjazz commited on Jun 6, 2024

alibi (#19)

ab85772
verified

jupyterjazz

Jackmin108 commited on Jun 4, 2024

lora bugfix (#16)

7c4a80c
verified

jupyterjazz

Jackmin108 commited on May 24, 2024

Update config.json (#15)

98c3cd2
verified

bwang0911 commited on May 21, 2024

Update tokenizer_config.json (#14)

c8da5f5
verified

bwang0911 commited on May 21, 2024

Update modeling_xlm_roberta.py (#13)

d230f23
verified

bwang0911 commited on May 20, 2024

Update modeling_xlm_roberta.py (#12)

64c81c6
verified

bwang0911 commited on May 20, 2024

truncate-embedding-dimension (#10)

8542ad8
verified

jupyterjazz commited on May 16, 2024

lora-multiple-adapters (#11)

27d23b2
verified

jupyterjazz commited on May 15, 2024

Refactor LoRA (#8)

4b000ec
verified

jupyterjazz commited on May 14, 2024

support lora (#1)

f9b3adb
verified

jupyterjazz commited on Apr 22, 2024

Support for SequenceClassification (#7)

0bb73e5
verified

michael-guenther commited on Apr 22, 2024

Support torch_dtype and CLS pooling (#6)

13c4251
verified

michael-guenther commited on Apr 19, 2024

Update modeling_xlm_roberta_for_glue.py

102e7bc
verified

koukandre commited on Apr 17, 2024

Create modeling_xlm_roberta_for_glue.py (#4)

4dd727d
verified

koukandre commited on Apr 17, 2024

add-encode-function (#5)

424df3c
verified

michael-guenther commited on Apr 17, 2024

support-cpu (#2)

290e593
verified

michael-guenther commited on Apr 17, 2024

Update configuration_xlm_roberta.py

807ba34
verified

jupyterjazz commited on Apr 16, 2024

Update configuration_xlm_roberta.py

9db6c6f
verified

jupyterjazz commited on Apr 16, 2024

add stochastic_depth

77af1c7

michael-guenther commited on Apr 15, 2024

support activation checkpointing

1c61b96

michael-guenther commited on Apr 12, 2024

add mlm model and adjust naming

95b4916

michael-guenther commited on Apr 11, 2024

add script to convert weights

eb21270

michael-guenther commited on Apr 10, 2024

add tokenizer class

3d87c79

michael-guenther commited on Apr 9, 2024

change config name

30e6a10

michael-guenther commited on Apr 9, 2024

upload model

2e3ebcb

michael-guenther commited on Apr 9, 2024

initial commit

2aec9c9
verified

michael-guenther commited on Apr 9, 2024

Commit History

fix: update frequencies when updating the rope base value (#40) 8b2ad1e verified

feat: update the readme a6bb16f

fix mixed precision loading with recent transformers versions (#39) bda5e8d verified

lora-instructions (#36) a709b51 verified

fix-device-issues (#38) 57dbe22 verified

fix: override use_flash_attn in lora dc4080e verified

Fix rotary backward 169b7fb verified

cpu-inference (#35) 80b1ff7 verified

feat: configurable use_reentrant (#37) 09dbf45 verified

refine-codebase (#33) 2646361 verified

fix-adapter-masks (#32) 0f0bed6 verified

change rotary base (#31) 4434bf3 verified

2-adapter-tuning (#29) 95fd08c verified

Update modeling_xlm_roberta.py 027219a verified

2-adapter-tuning-initial-impl (#30) 2cc8472 verified

Update modeling_xlm_roberta.py 3eceb33 verified

make-lora-stateless (#25) e860caa verified

feat-routing (#26) 6a92924 verified

fix-rope-inference (#28) 955fea2 verified

no-flash-attention-during-inference (#22) 7ad815b verified

Rename config to config.json 4d09ca8 verified

Create config 721e106 verified

Delete config.json a60f7c0 verified

rope-embeddings (#20) e3681c2 verified

alibi (#19) ab85772 verified

lora bugfix (#16) 7c4a80c verified

Update config.json (#15) 98c3cd2 verified

Update tokenizer_config.json (#14) c8da5f5 verified

Update modeling_xlm_roberta.py (#13) d230f23 verified

Update modeling_xlm_roberta.py (#12) 64c81c6 verified

truncate-embedding-dimension (#10) 8542ad8 verified

lora-multiple-adapters (#11) 27d23b2 verified

Refactor LoRA (#8) 4b000ec verified

support lora (#1) f9b3adb verified

Support for SequenceClassification (#7) 0bb73e5 verified

Support torch_dtype and CLS pooling (#6) 13c4251 verified

Update modeling_xlm_roberta_for_glue.py 102e7bc verified

Create modeling_xlm_roberta_for_glue.py (#4) 4dd727d verified

add-encode-function (#5) 424df3c verified

support-cpu (#2) 290e593 verified

Update configuration_xlm_roberta.py 807ba34 verified

Update configuration_xlm_roberta.py 9db6c6f verified

add stochastic_depth 77af1c7

support activation checkpointing 1c61b96

add mlm model and adjust naming 95b4916

add script to convert weights eb21270

add tokenizer class 3d87c79

change config name 30e6a10

upload model 2e3ebcb

initial commit 2aec9c9 verified

fix: update frequencies when updating the rope base value (#40)

8b2ad1e
verified

feat: update the readme

a6bb16f

fix mixed precision loading with recent transformers versions (#39)

bda5e8d
verified

lora-instructions (#36)

a709b51
verified

fix-device-issues (#38)

57dbe22
verified

fix: override use_flash_attn in lora

dc4080e
verified

Fix rotary backward

169b7fb
verified

cpu-inference (#35)

80b1ff7
verified

feat: configurable use_reentrant (#37)

09dbf45
verified

refine-codebase (#33)

2646361
verified

fix-adapter-masks (#32)

0f0bed6
verified

change rotary base (#31)

4434bf3
verified

2-adapter-tuning (#29)

95fd08c
verified

Update modeling_xlm_roberta.py

027219a
verified

2-adapter-tuning-initial-impl (#30)

2cc8472
verified

Update modeling_xlm_roberta.py

3eceb33
verified

make-lora-stateless (#25)

e860caa
verified

feat-routing (#26)

6a92924
verified

fix-rope-inference (#28)

955fea2
verified

no-flash-attention-during-inference (#22)

7ad815b
verified

Rename config to config.json

4d09ca8
verified

Create config

721e106
verified

Delete config.json

a60f7c0
verified

rope-embeddings (#20)

e3681c2
verified

alibi (#19)

ab85772
verified

lora bugfix (#16)

7c4a80c
verified

Update config.json (#15)

98c3cd2
verified

Update tokenizer_config.json (#14)

c8da5f5
verified

Update modeling_xlm_roberta.py (#13)

d230f23
verified

Update modeling_xlm_roberta.py (#12)

64c81c6
verified

truncate-embedding-dimension (#10)

8542ad8
verified

lora-multiple-adapters (#11)

27d23b2
verified

Refactor LoRA (#8)

4b000ec
verified

support lora (#1)

f9b3adb
verified

Support for SequenceClassification (#7)

0bb73e5
verified

Support torch_dtype and CLS pooling (#6)

13c4251
verified

Update modeling_xlm_roberta_for_glue.py

102e7bc
verified

Create modeling_xlm_roberta_for_glue.py (#4)

4dd727d
verified

add-encode-function (#5)

424df3c
verified

support-cpu (#2)

290e593
verified

Update configuration_xlm_roberta.py

807ba34
verified

Update configuration_xlm_roberta.py

9db6c6f
verified

add stochastic_depth

77af1c7

support activation checkpointing

1c61b96

add mlm model and adjust naming

95b4916

add script to convert weights

eb21270

add tokenizer class

3d87c79

change config name

30e6a10

upload model

2e3ebcb

initial commit

2aec9c9
verified