nanotron

View on PyPIReverse Dependencies (1)

0.4 nanotron-0.4-py3-none-any.whl

Wheel Details

Project: nanotron
Version: 0.4
Filename: nanotron-0.4-py3-none-any.whl
Download: [link]
Size: 163614
MD5: dc58efb584292e80525e4c2f1bbdec60
SHA256: 0c1834e91c17c651f430d46a1e779bc49991c33bda26cc237beae7d12b383ad9
Uploaded: 2024-03-05 10:21:29 +0000

dist-info

METADATA

Metadata-Version: 2.1
Name: nanotron
Version: 0.4
Summary: Minimalistic Large Language Model Training and Finetuning
Author-Email: Nouamane Tazi <nouamane[at]huggingface.co>, Thomas Wang <thomas.wang[at]huggingface.co>, Kunhao Zheng <kunhao[at]huggingface.co>, Thomas Wolf <thomas[at]huggingface.co>
Classifier: Topic :: Software Development
Requires-Python: ~=3.10
Requires-Dist: torch (>=1.13.1)
Requires-Dist: pyyaml
Requires-Dist: numpy
Requires-Dist: packaging
Requires-Dist: safetensors
Requires-Dist: dacite
Requires-Dist: tqdm
Requires-Dist: pre-commit; extra == "dev"
Requires-Dist: pylint; extra == "dev"
Requires-Dist: flash-attn (>=2.5.0); extra == "fast-modeling"
Requires-Dist: pytest; extra == "test"
Requires-Dist: pytest-xdist; extra == "test"
Provides-Extra: dev
Provides-Extra: fast-modeling
Provides-Extra: test
Description-Content-Type: text/markdown
License-File: LICENSE
[Description omitted; length: 4187 characters]

WHEEL

Wheel-Version: 1.0
Generator: bdist_wheel (0.42.0)
Root-Is-Purelib: true
Tag: py3-none-any

RECORD

Path Digest Size
nanotron/__init__.py sha256=XS7jokJxwt5Jz1m67HjCoJuIkOXxpZq0y16wzhrxpPU 20
nanotron/constants.py sha256=DB5cO2VtO5isOZ3aE7m4ilF3Vc-U-3YhDTYRduvBMbs 146
nanotron/dataloader.py sha256=WX7LKK6UsaeTnBDvtH5gt5OfRp6ggxZZp38cKOE3SdY 22600
nanotron/distributed.py sha256=feR5LHSSPBl7vATg1Hj8dVLQcsQYd9hbVcPu-RO5q9s 10600
nanotron/helpers.py sha256=4_LrxgHS7fM7C6B6yX-NQ8myuG8CGpVl_dCJzMTeRk8 20291
nanotron/logging.py sha256=H4z_Lm0R3LgtUSkBPThfovxGALs8bnW1gHarSkCUo_4 10143
nanotron/random.py sha256=dOJ0yGmB3BiY3cHDQDul7W8e7QT4zceolocOya7XC78 4700
nanotron/sanity_checks.py sha256=nFMTsJV-_lfFOlMuARHk3Mq__bJXMOtJR2WKbrF95oA 9805
nanotron/trainer.py sha256=RB3iVG86LEPajPwRFV4-J07KxIjuIuRzTbWmuQAm8jg 35249
nanotron/utils.py sha256=fToOh3BQbrQH_fsCUW37ZqNyqXoApqRDTKPEtyJ1Omw 5226
nanotron/config/__init__.py sha256=Xhj3aLhAazqW0vr3Ve7p7tcVXl4A09Zm-Nf8aPwu7TA 186
nanotron/config/config.py sha256=dTWKnUhX2Y4OP7gwMRcO3VZfd3pOFev92x8Xq09iVIM 13953
nanotron/config/lighteval_config.py sha256=BCwDpDucUs-rMFTT5g2MnAyXdQtOfKjCmKIqc7q5W8w 2933
nanotron/config/models_config.py sha256=OoyDkF5mYTdmtsXBmBUYkOdjEq8ex-nmPK1SjXPfujc 3543
nanotron/config/parallelism_config.py sha256=CG5OXU704vilF56oCtRgG5MF56GqlRzSOg4ncD58NEo 1752
nanotron/config/utils_config.py sha256=R0L0tiyJuoKnKY7ojh2Bv6DP2rQyjyuvQH4kYiMoVj4 3612
nanotron/fp8/__init__.py sha256=kNNaMxeLtqdSPLBFUztvdnSdLpqvDgwiO87qLQwdIBw 419
nanotron/fp8/constants.py sha256=Meby-gLCxOApcMDX5uPoU4U4y19FuZEa2dgdjv9Rkao 441
nanotron/fp8/dtypes.py sha256=Tov3hkiaao1KDCo83WC7YUXHaypAXqSD26r43_uxkIo 114
nanotron/fp8/kernel.py sha256=3_4owTaLnOPeimRJycDDK3ikcoXN3nJ1nOfeWGM2OcE 2057
nanotron/fp8/linear.py sha256=zG4P_M6K2to5hjUcr4AlcxhAH-0WiIn7gLv4eykgaUQ 4781
nanotron/fp8/meta.py sha256=V5VwZUdygJ5jHPjx_N1cmSdZ8up91zv9YBTJzYek6QQ 1198
nanotron/fp8/parameter.py sha256=ogIwi-iBdDBUMUYyRmZdgmGEau2XyY4eQ3aXiyCXv5k 1632
nanotron/fp8/tensor.py sha256=AGaTMinofW9f6SB-qS7o-aXdwuVrWTz86Q92wIO6NTk 4699
nanotron/fp8/utils.py sha256=RaX1qjTdfuGESoJ2KmD5etIanAYcfj2a_-2l6gtHAbA 421
nanotron/generation/__init__.py sha256=OarEPG2G_RDEsgepHESjnCEnXNm9LwBlnsCI4KzO1cA 198
nanotron/generation/decode.py sha256=rvxYW8X88-KsRBnRjGdRHM-kYPsu0YQTNjMBN_H8Y_E 40888
nanotron/generation/generate_store.py sha256=liHDVxfr_aiIDC0tEhs1torP2DXLIL_DjY5ADSXBxJc 1550
nanotron/generation/sampler.py sha256=wVKRWEnXEwI8cj7VwlmROcVwzfJOinGff3IkqyJ--g0 13758
nanotron/models/__init__.py sha256=EDAZQjt3yp1aYMV7c25VyuelEvQp-1wiB1xT6oFAj04 130
nanotron/models/base.py sha256=TxwRNhKd94v8QZ56LkfrYkft32qhjTCbJWyHt7VdhxY 11478
nanotron/models/llama.py sha256=EJlsWyGIqJLTo1xrpVEfNWO9nEwtDTocJ77Gipv1eOs 45742
nanotron/models/starcoder2.py sha256=MeOHMpbHdZipbkQJoj3kXqzLfS_j_CEBP1qiFF_fuCk 71845
nanotron/nn/__init__.py sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU 0
nanotron/nn/activations.py sha256=nIlj34cD9i16eOFBNIMqi0UjBbAhQBjXN4iEw6PZ0_M 8784
nanotron/nn/layer_norm.py sha256=km-25m4KWL7ayel8cJEi6l1yDFstdu-CE_CJBGa7RHg 1568
nanotron/optim/__init__.py sha256=JCYDRwV2S87-csom079HbKOBHFxCI5vuWCESgepTCN0 502
nanotron/optim/base.py sha256=2-7sm5qmXJIjH7Z0FQ9T3pIPheViJhlKQcW1WVa2Tio 1087
nanotron/optim/clip_grads.py sha256=9LzUWr0AJcH5FmTE1FeSb7wXvox4vY2RArbcklLEHWk 4147
nanotron/optim/gradient_accumulator.py sha256=JGwhs9G_K4Rb9_GgTWBsSTvd59IdSYtsKGnVvEQyBOw 16937
nanotron/optim/inherit_from_other_optimizer.py sha256=KwKXORwFP4nqwAE8vJllO-7HmGLn5kvTe-jnbJSm3Lc 1754
nanotron/optim/named_optimizer.py sha256=1PTGewa4fVJWg56Tz89GbtvU1BHvlPnrPWEpMbIyfkI 2936
nanotron/optim/optimizer_from_gradient_accumulator.py sha256=VLuO-jQUlMn9iYGzBBb0F1J3Vqi1MuBo6V2858zzNgo 3104
nanotron/optim/zero.py sha256=0AB14nJI81IRetox-f1fy20g4bT1JIth_ky2Gr_94gM 23065
nanotron/parallel/__init__.py sha256=s-UMh8WBjiAfSbxRgVHHL00EKeJN1eeWVXWCq0jsSX8 69
nanotron/parallel/context.py sha256=h49jMmffnlTpwM-9Z7tNLsL3mvrxsnmpjCiLJhEbWvI 4955
nanotron/parallel/parameters.py sha256=XGbPecgYyooboR8_f7uCtGZn_fl9XyRETemPOIPnmx0 7767
nanotron/parallel/sharded_parameters.py sha256=rsMbn8OlIn_YZ1IEMxNrMTM7Q1G3UVBiZZqYWrPe9DI 5157
nanotron/parallel/tied_parameters.py sha256=yVpYZah2asV7RBBG8Dl50KGeGnRReI4nh2NPw4F5e94 6551
nanotron/parallel/utils.py sha256=PtTehpauzJ7HdZBDJY_qYoMZoT6XhuULprj1KQxS74w 1247
nanotron/parallel/data_parallel/utils.py sha256=BRwMZiePda9lIxAb9U2ovQSgCyy3EX4IH7FqTSSzmp0 1968
nanotron/parallel/pipeline_parallel/__init__.py sha256=g2ik5BMYudbdE2BVDGM64DkLXyoykwcYgf8tJFwyLy4 281
nanotron/parallel/pipeline_parallel/block.py sha256=UBryA5H0E3_Zj3mzQmp-nUnyd0jbnl5h9Yr0xYP33OM 8728
nanotron/parallel/pipeline_parallel/context_manager.py sha256=7NQweGnxETknCYuCS4T-Pp-wLMkvZT6q3GVd-az7X7I 1013
nanotron/parallel/pipeline_parallel/engine.py sha256=rw0Mdbk96WeuaOX0awu8Uo99NwJlAlHtp15ytjGVqmc 14227
nanotron/parallel/pipeline_parallel/functional.py sha256=gwzVeYypQskKM17EFDIoMBAB9VliF-XDZA_n_NJSWWY 4190
nanotron/parallel/pipeline_parallel/p2p.py sha256=vCf4Ta0fIah2udntB2Vzwz9GV_RYMekxMjKgZOf8nXM 20171
nanotron/parallel/pipeline_parallel/state.py sha256=xmynF8uyAS5r6g1uYHS7JOvg_LMYWDiv1kDxuXB4VFI 12439
nanotron/parallel/pipeline_parallel/tensor_pointer.py sha256=CgNxMlsCUCeB9HjkTf_ef3GriocP4PCw8jmp6UjBeo4 452
nanotron/parallel/pipeline_parallel/utils.py sha256=QgoWFN4Gcf5Umn3wz4vAzo4GS5loM8SGs-TviO4CDxk 1407
nanotron/parallel/tensor_parallel/__init__.py sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU 0
nanotron/parallel/tensor_parallel/distributed_differentiable_primitives.py sha256=4-Ug_bnpyFhx8sYw-fxNQv71DkHtlw36RSuzGX_dJqY 4796
nanotron/parallel/tensor_parallel/enum.py sha256=7CVJV5B-4s5Mt7mOCn4fDzMiA6HXoTFlkTNvZvStzCo 206
nanotron/parallel/tensor_parallel/functional.py sha256=F_qhBXRcfHS0oY-FnfTOAYeXdS7-y0TKAqiPkVky4wA 20601
nanotron/parallel/tensor_parallel/nn.py sha256=Ffda-gg3MMKGZlOIA1c2kQPe7V9_Kz74PjpBGMqADlQ 10295
nanotron/serialize/__init__.py sha256=nlZfihiwtvm5S6GFXWMrhIojHLj-XR1omtY2SgYW2eA 136
nanotron/serialize/main.py sha256=A6gzAkeZyF1JFkk_nU-SMDiAMuHUvG2_K2TUbTZrHgw 10564
nanotron/serialize/metadata.py sha256=r3n_7usMRW0IasgojyXTvyrYcoTEkmaQl8gJMni6eFU 4182
nanotron/serialize/optimizer.py sha256=K3HM4E1LKIbbCQJvh9cz50wz9D-c4UEWuNOQ-BAleg8 17066
nanotron/serialize/random.py sha256=lcGnjuZCk6xbtr9-5FN77XOrUHuUqL3ugq0ADAjzy4k 1551
nanotron/serialize/utils.py sha256=BS0mXu8qErcc8cNQnR62nB60wZp69suOUCfPvZTAk2o 2382
nanotron/serialize/weights.py sha256=tAfHj8SpH3L3mUVbD6BGptIHcHMhQicOmuruk18mzms 17127
nanotron-0.4.dist-info/LICENSE sha256=xx0jnfkXJvxRnG63LTGOxlggYnIysveWIZ6H3PNdCrQ 11357
nanotron-0.4.dist-info/METADATA sha256=-sZCkhkCX4pamZZzB4P-0SCJ-XjfQIHiGEihQQajx3c 5083
nanotron-0.4.dist-info/WHEEL sha256=oiQVh_5PnQM0E3gPdiz09WCNmwiHDMaGer_elqB3coM 92
nanotron-0.4.dist-info/top_level.txt sha256=4LTLLbCZoHkoAVCQ6jIC5qUFyQKoUjWJYGW23pXW40w 9
nanotron-0.4.dist-info/RECORD

top_level.txt

nanotron