Wheelodex » Projects » attention-sinks

attention-sinks

View on PyPI — Reverse Dependencies (2)

0.4.0 attention_sinks-0.4.0-py3-none-any.whl

Wheel Details

Project: attention-sinks
Version: 0.4.0
Filename: attention_sinks-0.4.0-py3-none-any.whl
Download: [link]
Size: 35799
MD5: d15c559b080fa53112bc788664424313
SHA256: e305b536376dab8c2cf40ee363805c54d7949a720c7bf6e5fb0fc825e0b127f2
Uploaded: 2023-11-23 12:11:01 +0000

dist-info

METADATA · WHEEL · RECORD · top_level.txt

METADATA

Metadata-Version: 2.1
Name: attention-sinks
Version: 0.4.0
Summary: Extend LLMs to infinite length without sacrificing efficiency and performance, without retraining
Author: Tom Aarsen
Maintainer: Tom Aarsen
Project-Url: Repository, https://github.com/tomaarsen/attention_sinks
License: Apache-2.0
Keywords: data-science,natural-language-processing,artificial-intelligence,mlops,nlp,machine-learning,transformers
Requires-Python: >=3.8
Requires-Dist: torch
Requires-Dist: transformers (==4.34.0)
Requires-Dist: tokenizers (<0.15,>=0.14)
Requires-Dist: pre-commit; extra == "dev"
Requires-Dist: ruff; extra == "dev"
Requires-Dist: black; extra == "dev"
Requires-Dist: pytest; extra == "dev"
Requires-Dist: pytest-cov; extra == "dev"
Requires-Dist: spacy; extra == "dev"
Provides-Extra: dev
Description-Content-Type: text/markdown
License-File: LICENSE
[Description omitted; length: 19736 characters]

WHEEL

Wheel-Version: 1.0
Generator: bdist_wheel (0.41.2)
Root-Is-Purelib: true
Tag: py3-none-any

RECORD

Path Digest Size
attention_sinks/__init__.py sha256=bf57vNNlxOS2L4u7HF2tOGvJokyfu54Wfie4GiBEO2M 1117
attention_sinks/attention_sink_kv_cache.py sha256=80VkjX2YfbrMq0HtMqUxJtt1voSAzzQCyrQGeQaFiZQ 3799
attention_sinks/inject_mixin.py sha256=52ZzlxktRq945ijKnyk0QY0z3Pka33bvrzjD4lIj8Y4 6022
attention_sinks/generation/utils.py sha256=26ih5sAfl96XNLDivCnEXYbWnDe4kUeLmpdguddnpbo 2153
attention_sinks/models/__init__.py sha256=XqjgZe52oEmuK7TywGwdBry09tXDmF06pCOKnU8HLro 1491
attention_sinks/models/auto/__init__.py sha256=o429Ny3pregJw8LEmiG0Uzocbefk278Ymv3mYL6IxBM 191
attention_sinks/models/auto/modeling_auto.py sha256=s-sRrfINyJ-z9JfTbxn3lB5ay2WUoOgQhxadDLn8NTQ 987
attention_sinks/models/falcon/__init__.py sha256=KE8uj8a10VqBj1KG_HiKvbkzQueUqM7yTzFh8ZyWGNs 270
attention_sinks/models/falcon/modeling_falcon.py sha256=M15WTFQc2oPo7b53oKDKP_BCxOZPY7m_6mzZZspvatg 1096
attention_sinks/models/falcon/pos_shift.py sha256=BOQhb5z1jAG3m4fcBxvxPRrg9iKGFku7yAnJuYzy2zI 6421
attention_sinks/models/gpt_neox/__init__.py sha256=OMCouNDfjxf1plrz9qDpcEIOx4ciHVYV2Y2w_cCcjqs 280
attention_sinks/models/gpt_neox/modeling_gpt_neox.py sha256=6XrII-p_G9fgdxoogrE-TDAC7fSLHWfY_KVzkm_hGM8 1125
attention_sinks/models/gpt_neox/pos_shift.py sha256=AQ43Cgq9Bow35CBajzfH7wvyQd47AqXWr7ObeHq5G40 3373
attention_sinks/models/gptj/__init__.py sha256=C3pVSvkSLmkBlFFp8uuzJO_cgpUyFfZzPOhDzrcz50o 221
attention_sinks/models/gptj/modeling_gptj.py sha256=qwKoZa-xKuZXg-6NGTeruB_sAgHi78ccdKjSRFClP1Q 855
attention_sinks/models/gptj/pos_shift.py sha256=QhDKiFlsUUQNUCe5pUxDyP17gDLCpYAg_cgXaeJdu8g 4009
attention_sinks/models/llama/__init__.py sha256=UDMv7XnQdUO6L3imA5YH3IslzfLVzPIhv8qhJLZoorI 148
attention_sinks/models/llama/modeling_llama.py sha256=_OgFB_1C1B07H1wy6zKZq_fvp90BpO3Qv9Pdxgm71RU 699
attention_sinks/models/llama/pos_shift.py sha256=vZyrVsOvkHfmIQmjK_4eebezjQVzbIdqelgY-tUl5lE 5670
attention_sinks/models/mistral/__init__.py sha256=0gL91b-OClbqBJEfaQfWJI8J7HO-wyLXZbx6k0LMcNU 179
attention_sinks/models/mistral/modeling_mistral.py sha256=sRdeDldn6gGr-QX35JLafgK1uX4MC91qNwAgjRAaC1M 737
attention_sinks/models/mistral/pos_shift.py sha256=1NOxn3iAwiL8e8z3xuPwuNsGpvfX6o5z5brzxhvSw5A 4267
attention_sinks/models/mpt/__init__.py sha256=mPhmgb105AaHxz1WrwnOnFIMCXhdUwMGmtxcQd0rMaY 188
attention_sinks/models/mpt/modeling_mpt.py sha256=hatTlVrAO7byHh6S3itdBPfAFcGLY_Q0MmWZexojn9I 1009
attention_sinks/models/qwen/__init__.py sha256=Xy6cRhiQVw04A5A9YXGo2cX9r_V3oLyQwVax9w4rHl0 57
attention_sinks/models/qwen/pos_shift.py sha256=jvOvFGEq9JVks3HaUNcTuPzhxZtK58BIu2MtuyeHSy8 4311
attention_sinks/models/stablelm_epoch/__init__.py sha256=HbAcrLeb_643u5tTK7swkqfoxbaEJO5EEQ_vrpI5IWM 65
attention_sinks/models/stablelm_epoch/pos_shift.py sha256=R988kRjP1fw55NsFYMl0aVwB144YIh29QDC4g5Cr0So 5286
attention_sinks/models/yi/__init__.py sha256=P-VrJ4UKEoysBPv1O_gpxh0uNv6jA7ogAH4A97J8nHw 53
attention_sinks/models/yi/pos_shift.py sha256=YYAtNG8QSlnBgNHxGNI6HD7SitQ8ugN-7iREOHqUuxk 5979
attention_sinks-0.4.0.dist-info/LICENSE sha256=HrhfyXIkWY2tGFK11kg7vPCqhgh5DcxleloqdhrpyMY 11558
attention_sinks-0.4.0.dist-info/METADATA sha256=sLHTRGEHN8qdwTnj3IKv9qx39zNPVpJixqgSGK6y4fo 20962
attention_sinks-0.4.0.dist-info/WHEEL sha256=yQN5g4mg4AybRjkgi-9yy4iQEFibGQmlz78Pik5Or-A 92
attention_sinks-0.4.0.dist-info/top_level.txt sha256=4HcJFFllFRtxgOMDV_hpVZ79cIG3Tp9VIITSDpDKUzM 16
attention_sinks-0.4.0.dist-info/RECORD — —

top_level.txt

attention_sinks