Wing Lian 87455e7f32 swaps to use newer sample packing for mistral (#1773) ...

* swaps to use newer sample packing for mistral

* fix multipack patch test

* patch the common fa utils

* update for refactor of flash attn unpad

* remove un-needed drop attn mask for mistral

* bump transformers to main to pick up latest mistral fix for 12b and refactor of fa2

* update test

2024-07-23 01:41:11 -04:00

..

__init__.py

attempt to also run e2e tests that needs gpus (#1070)

2024-01-09 21:23:23 -05:00

test_4d_multipack_llama.py

support for true batches with multipack (#1230)

2024-02-01 10:18:42 -05:00

test_fa_xentropy.py

bump flash attention 2.5.8 -> 2.6.1 (#1738)

2024-07-14 19:11:31 -04:00

test_falcon_samplepack.py

Falcon embeddings (#1149) [skip docker]

2024-01-22 21:01:42 -05:00

test_fused_llama.py

make mlflow optional (#1317)

2024-02-26 11:41:33 -05:00

test_llama_s2_attention.py

support for gemma2 w sample packing (#1718)

2024-06-29 01:38:55 -04:00

test_lora_llama_multipack.py

more fixes 20240228 (#1342) [skip ci]

2024-02-28 12:57:45 -05:00

test_mistral_samplepack.py

relora: magnitude pruning of the optimizer (#1245)

2024-02-06 00:35:30 -05:00

test_mixtral_samplepack.py

fix broken linting (#1541)

2024-04-19 01:03:04 -04:00

test_model_patches.py

swaps to use newer sample packing for mistral (#1773)

2024-07-23 01:41:11 -04:00

test_phi_multipack.py

Phi2 multipack (#1173)

2024-01-23 12:54:36 -05:00

test_resume.py

make sure the CI fails when pytest script fails (#1669)

2024-05-29 10:12:11 -04:00

test_unsloth_integration.py

add tests so CI can catch updates where patches will break with unsloth (#1737) [skip ci]

2024-07-11 16:43:19 -04:00