Add files using upload-large-folder tool

README.md

@@ -1,3 +1,39 @@
----
-license: mit
----
+# Linear Next Benchmark
+
+Linear Next is a comprehensive benchmark designed to fairly compare various efficient transformer architectures. This project evaluates different approaches including linear attention, sparse attention, and other model structures under identical training conditions and datasets.
+
+## Overview
+
+The benchmark aims to provide an unbiased comparison of efficient transformer variants by ensuring all models are trained with the same datasets, hyperparameters, and evaluation metrics. This allows for a clear understanding of the relative strengths and weaknesses of each approach.
+
+## Datasets
+
+The benchmark utilizes a diverse collection of high-quality datasets:
+
+### General Text
+- **DCLM-pro**: A large-scale dataset containing diverse text from various domains, designed for general language modeling tasks.
+- **Cosmopedia-v2**: A curated corpus of high-quality web content covering a wide range of topics, with emphasis on educational and informative material.
+- **Fineweb-edu**: A filtered collection of educational web content, focusing on instructional and academic text from reliable sources.
+
+### Code
+- **The Stack v2**: A comprehensive collection of source code spanning multiple programming languages, designed to train models on code understanding and generation tasks.
+
+### Mathematics
+- **Finemath**: A specialized dataset containing mathematical content, including equations, proofs, and mathematical explanations across various difficulty levels.
+
+### Reasoning
+- **Natural Reasoning**: A dataset focused on logical reasoning, problem-solving, and inference tasks, designed to improve models' reasoning capabilities.
+
+## Methodology
+
+All models in the Linear Next benchmark are evaluated using identical:
+- Training datasets and data mixing ratios
+- Optimization parameters
+- Hardware configurations
+- Evaluation metrics
+
+This controlled environment ensures that performance differences can be attributed to the architectural differences rather than training conditions.
+
+## Results
+
+Detailed benchmark results, including training curves, inference speed, memory usage, and performance metrics across different tasks, are available in the project repository.

code/code-part-00_of_02/train-00004-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:365cdb9d1108c319d7d42e24e43f9736087c271ddc87b4a68a68f61381819fb0
+size 357204185

code/code-part-00_of_02/train-00018-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e01d4fc9dd2891fe2fb54e6f6a63e43a0e27c99e62407a15e2b726711d4e35e9
+size 358835555

code/code-part-00_of_02/train-00029-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2a73e3d4079fce29f25ac246acd4201420c89df78ded5fcfff7be17696752949
+size 355384566

code/code-part-00_of_02/train-00032-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3c020324800ac971be0eb1481ad83dd5c9e12f220adc7e0ab5ba2404bbdaad28
+size 353431125

code/code-part-00_of_02/train-00042-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c7470c3904d802f21b972a2d28219e68de7d681f55cd23bf48c7db88a125e6a
+size 358370453

code/code-part-00_of_02/train-00043-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f75bcaea004aeaf86b364f97cf34940d198dc141e6ec3762860b2554ee4d726
+size 351990837

code/code-part-00_of_02/train-00044-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b704757ef0f25d01b1e8815c0ebbd91d249be105e9e871c1599fef16edd6729
+size 356224962

code/code-part-00_of_02/train-00054-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f49c1e14924c146cc730ef55ac4a6876a3a979f212eeac710529b709b2f1ff9e
+size 355352022

code/code-part-00_of_02/train-00062-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd020521d45c493efdc567db041b2871534f2d154b8bb7b9972102ad4533d84b
+size 364798388

code/code-part-00_of_02/train-00075-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:82777fd2356e47fa37bae10f4c3a88081098ec8674e4380b3989d143a9b8095f
+size 356496867

code/code-part-00_of_02/train-00078-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0ab1824f8cccbaeaa1676346099df1d9eef25500d30dc52e363f4f1083be9298
+size 352498261

code/code-part-00_of_02/train-00099-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02ec621ba4804365e4131fe49f737f8fdad152bbea321df75355c7292dade155
+size 355780466

code/code-part-00_of_02/train-00107-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7ca4bbd29953704be8c35cf4759d2bd0803517ef5d3e848863a42ae01f2736c8
+size 352750799

code/code-part-00_of_02/train-00111-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cbf1f2b181f5c2c6063ad3213fe8b480d4ebae642ecb520e494107ae7352385c
+size 367985134

code/code-part-00_of_02/train-00112-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8edaeec599774b9da5a133a8fb64448092e29a60d5614ffd9bd6777d6efaceba
+size 353748910

code/code-part-00_of_02/train-00113-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:417f9b8e153e23827bee36bd473ca8364d7ea6fcfcf1762a13e44826d3dcc802
+size 355635812

code/code-part-00_of_02/train-00115-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:13e2cdc7871aa1abb40d61cf525963627cbef9f81303253a330afb36ab3b6a93
+size 354323080

code/code-part-00_of_02/train-00116-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:552ec9721900463bd5e2184c5f02e2539d631424261bce580190fab52ae32958
+size 356083248

code/code-part-00_of_02/train-00122-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40d300d912c5148800f3eeb6d6c58da27fcfbbe079157573146897fe86a7fc83
+size 355076754

code/code-part-00_of_02/train-00125-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:02bca6a34f9609d979c51e28ef2d4c7c79c23b21cce0a6799e1f9ac718a287cf
+size 355866779

code/code-part-00_of_02/train-00136-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5466678ed5145dddd9a571e13196360053a8a7bf8bee73f2c209b68e9db7b80b
+size 357055327

code/code-part-00_of_02/train-00146-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:729cb94bcda0a6fc4eb22e3b1168727f415c07ccf137b021c511934e1f564a47
+size 356947075

code/code-part-00_of_02/train-00148-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c45c8e9ae61366c3b70f732c6f887d876846eb377006267fdfa9d13e42187d4
+size 358067882

code/code-part-00_of_02/train-00156-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:46f957243da14d97c141f481a72658f403c0fe692e18e964b91658f9a492c63c
+size 353201312

code/code-part-00_of_02/train-00159-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:250ab126aff5d6168ddb29425baad0b048ebb16d2b3bd1d86ec1e466eef79b99
+size 352761541

code/code-part-00_of_02/train-00173-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd579247b3d35dd979cd31a5d4458d15b8fa0832ffdd48768edad4c7f69e8815
+size 357827578

code/code-part-00_of_02/train-00183-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87a7b2def5d2d69be69e393834745a02cab026a84cbede0c8c94a2693fe06120
+size 356636171

code/code-part-00_of_02/train-00187-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e02033bd242b1f5242b999a2df438a739e55401528a623bbd0231a6004902c40
+size 367130675

code/code-part-00_of_02/train-00202-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eca58a34767cd2e4f0e36cb81402a970c89ecb93e5fc9601f887f76f319916a
+size 355146117

code/code-part-00_of_02/train-00212-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6cd557d7d76719fcf61b6799b8ef83b35c7b8d7b767dfa0887cbd9f4137f39b7
+size 357820222

code/code-part-00_of_02/train-00219-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:386f996725fd35dac6d65e90b20a3b21b238d4d81a3672888567985e198fe62b
+size 353897840

code/code-part-00_of_02/train-00221-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b35e46937f20ecbb96f285f2d7e162d99dbe3cf154473d2f735f7362ed9ec9e2
+size 354227561

code/code-part-00_of_02/train-00222-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f95efa00c440a483d43c6a2e06ba6376d62ad4b5d21e914189b178af8d87ca2e
+size 358898223

code/code-part-00_of_02/train-00228-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3518aedc6b8b716dbe0643c458411cc395bae00fa195a08f8b95997bbd7a106c
+size 362084235

code/code-part-00_of_02/train-00229-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2629d3a87db215f9f02725aecc77be2c82db6582b3155720c8849666eca46d51
+size 354020923

code/code-part-00_of_02/train-00240-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a33c0be6baab84019ff0a17810fbe126f927a2c1b43394048d430c2eb092febb
+size 358998490

code/code-part-00_of_02/train-00254-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5f36964aa7c1b817a757443bac96db5dce13781e7ea0605d54fdf20996eba4da
+size 359526057

code/code-part-00_of_02/train-00262-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b63d62ff4db0d425017e50517f3ceb2b52208326b6327f870729eb8642db1614
+size 358416195

code/code-part-00_of_02/train-00264-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e1527eca1dbd07677feb06696aee6734bb0cb682395c9e08ee91d69f50b3a65
+size 354618412

code/code-part-00_of_02/train-00284-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:81f56b2bb17f9b0cc081661705cdc142ed26ffe2a406200d99c791b6bf9fb1a7
+size 351018043

code/code-part-00_of_02/train-00291-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:41069e6c22f23ce8538e0cc242891042b833d2d0440ba553899fa95a2ceaffbd
+size 357459291

code/code-part-00_of_02/train-00294-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e4b44d3ead204c066b7fa0732b3cb9d9b9cd51ff5de93e255379dbadf10162cb
+size 358473006

code/code-part-00_of_02/train-00296-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:60a1ded8e61629c5ad0f63025df162db3246d2c624ee36c1458ebdff4ebbd5a1
+size 355920709

code/code-part-00_of_02/train-00306-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:97aab3d2bf9820bf78326ad4620d813ed46cdf6d565e2def1a4cbaa1cd5cb92f
+size 361241777

code/code-part-00_of_02/train-00313-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eae3ef6186879deac5f05d5b67cd1b48cadfdcb020f5bec010c012d7dd7f17c6
+size 361487361

code/code-part-00_of_02/train-00314-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:32d062a3f5f9f4749b6fa8e311ceeb7d0cf93c7d9fcffcfd53dfc4491d1d445b
+size 356401297

code/code-part-00_of_02/train-00343-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1af42647122d2005553b6e022ec3be2f2781eba143212798ba61437a31fb830f
+size 355504203

code/code-part-00_of_02/train-00345-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c87c45b4f39bcae4182c90335a97edcdf47a0e4ab7e1d26e2f8ee2119971ff18
+size 355909167

code/code-part-00_of_02/train-00352-of-01024.parquet

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a77fff983190b29a66a542e1c950b9caa3e2ac81529c27663c77ff7851d8d755
+size 356483278