Skip to page content
SAVRN
Search Plan a project

Dataset

l1-exact-qwen3-1.7b-compression-1k-8k-alpha3e-4-bs32-n8-32k-t1-146102-rollouts

by Jiahao Zhang hi-todayis-jh/l1-exact-qwen3-1.7b-compression-1k-8k-alpha3e-4-bs32-n8-32k-t1-146102-rollouts

l1exactQwen3-1.7Bcompressionn1000to8000alpha3e-4bs32n832k1epoch One verified gzip JSONL shard per training step; 256 responses per shard. L1-Exact: MathVerify correctness after thinking without an EOS gate, plus the original linear target-length reward.

Rows—
Configurations—
Size316.5 MB
Licenseapache-2.0
AccessPublicly accessible
Monthly Downloads—

Dataset Card

By Jiahao Zhang, published under apache-2.0, revision d2253061958c.

RL training rollouts

l1_exact_Qwen3-1.7B_compression_n1000to8000_alpha3e-4_bs32_n8_32k_1epoch

One verified gzip JSONL shard per training step; 256 responses per shard. L1-Exact: MathVerify correctness after thinking without an EOS gate, plus the original linear target-length reward. Fixed random targets 1000..8000; alpha=0.0003.

Details

Repository
hi-todayis-jh/l1-exact-qwen3-1.7b-compression-1k-8k-alpha3e-4-bs32-n8-32k-t1-146102-rollouts
Publisher
Jiahao Zhang
Task category
Not stated by the source
Tags
Not stated by the source
Size category
Not stated by the source
Languages
Not stated by the source
Revision
d2253061958c26d97f0f480f52d3a982e4afbcee
Last updated
2026-10-05

Files

90 files, 316.5 MB in total.

Data88 files · 316.5 MB
Documentation1 file · 363 B
Repository1 file · 2.5 KB
Every file
FileTypeSizeSHA-256
data/step_000001.jsonl.gzData3.0 MB50e4e4829bbe
data/step_000002.jsonl.gzData3.3 MBc552d8320015
data/step_000003.jsonl.gzData3.6 MB4d9a50ce4513
data/step_000004.jsonl.gzData3.2 MB1a5ba9ad3a5c
data/step_000005.jsonl.gzData2.5 MB4592d3226f03
data/step_000006.jsonl.gzData3.2 MBfa04bdc5b611
data/step_000007.jsonl.gzData2.5 MB3392d3e66f66
data/step_000008.jsonl.gzData4.0 MBe3e704a5de69
data/step_000009.jsonl.gzData3.5 MB180ba45bc4d9
data/step_000010.jsonl.gzData3.9 MB9e420ab6cd04
data/step_000011.jsonl.gzData2.9 MBf409e9084be7
data/step_000012.jsonl.gzData2.6 MB6819c7129b06
data/step_000013.jsonl.gzData3.6 MB6fbe8a32347a
data/step_000014.jsonl.gzData2.9 MB15153d332ba0
data/step_000015.jsonl.gzData3.3 MB0d4e77a6e83e
data/step_000016.jsonl.gzData2.8 MB60498b349632
data/step_000017.jsonl.gzData3.3 MB0e7898e60cf7
data/step_000018.jsonl.gzData3.8 MB3f41aaf04b7c
data/step_000019.jsonl.gzData3.0 MBb82090d224af
data/step_000020.jsonl.gzData3.4 MB0739bce6a0a7
data/step_000021.jsonl.gzData3.4 MB2a0ce6a44710
data/step_000022.jsonl.gzData3.3 MBf7bf94d722e0
data/step_000023.jsonl.gzData4.1 MBbe3c39b1cba7
data/step_000024.jsonl.gzData3.4 MB995c627aba2a
data/step_000025.jsonl.gzData2.9 MB5ec9d1c944d6
data/step_000026.jsonl.gzData3.5 MB0c196d3b79b1
data/step_000027.jsonl.gzData3.1 MB995b039768a3
data/step_000028.jsonl.gzData3.7 MB60d8f552951d
data/step_000029.jsonl.gzData3.2 MBfc9f318d129a
data/step_000030.jsonl.gzData3.3 MBbaf553b612c2
data/step_000031.jsonl.gzData3.3 MBb817891fbb4c
data/step_000032.jsonl.gzData2.8 MBe84a722473ff
data/step_000033.jsonl.gzData2.8 MB16a969c35781
data/step_000034.jsonl.gzData3.9 MBbb1ff8fea7fb
data/step_000035.jsonl.gzData3.1 MBf27fe5d4a5c6
data/step_000036.jsonl.gzData3.0 MB20e6aeadd293
data/step_000037.jsonl.gzData3.1 MBba14973e78f7
data/step_000038.jsonl.gzData3.3 MB111f1127ae0f
data/step_000039.jsonl.gzData3.4 MB6c61312049a4
data/step_000040.jsonl.gzData3.2 MBbf36acfcd22d
data/step_000041.jsonl.gzData3.9 MB4337e1f2e167
data/step_000042.jsonl.gzData4.2 MB10405629a7dc
data/step_000043.jsonl.gzData3.4 MB5e74b8b15191
data/step_000044.jsonl.gzData3.5 MBe0eb91680229
data/step_000045.jsonl.gzData4.4 MBe933fe4eef57
data/step_000046.jsonl.gzData3.8 MB6b4481a58475
data/step_000047.jsonl.gzData3.7 MBc975b014cb09
data/step_000048.jsonl.gzData3.8 MB46bdb20d5053
data/step_000049.jsonl.gzData4.0 MB85fb1f6657ce
data/step_000050.jsonl.gzData4.2 MB469ba38862e9
data/step_000051.jsonl.gzData4.2 MB4950cebca214
data/step_000052.jsonl.gzData3.7 MB9341fef067b7
data/step_000053.jsonl.gzData3.7 MB5340515737c3
data/step_000054.jsonl.gzData3.4 MB2e0215899e2e
data/step_000055.jsonl.gzData4.2 MBde9a1dabea59
data/step_000056.jsonl.gzData3.1 MB99d561d8c6c0
data/step_000057.jsonl.gzData3.9 MB12341fbe0119
data/step_000058.jsonl.gzData3.9 MB0c619e49605a
data/step_000059.jsonl.gzData3.6 MBd833e98478ae
data/step_000060.jsonl.gzData4.1 MB3d1c4f976869
data/step_000061.jsonl.gzData3.9 MBfd3ffc741601
data/step_000062.jsonl.gzData4.3 MB8d7f968230cc
data/step_000063.jsonl.gzData3.8 MB4d76ef7bd024
data/step_000064.jsonl.gzData4.0 MB0adf2ee1c044
data/step_000065.jsonl.gzData4.5 MBe7bb1220e081
data/step_000066.jsonl.gzData4.1 MBc5ed400d4c08
data/step_000067.jsonl.gzData4.0 MB61cd99d77b47
data/step_000068.jsonl.gzData4.0 MB28e1d882c412
data/step_000069.jsonl.gzData3.7 MBe3ab7a389fa2
data/step_000070.jsonl.gzData3.9 MB494a66961a65
data/step_000071.jsonl.gzData4.2 MB5f52bc41e05e
data/step_000072.jsonl.gzData3.6 MBd005af143890
data/step_000073.jsonl.gzData3.8 MBafa517f65683
data/step_000074.jsonl.gzData4.5 MB29d1d86f0b54
data/step_000075.jsonl.gzData3.9 MBd972ffa3d08c
data/step_000076.jsonl.gzData3.7 MB83fb499a86c6
data/step_000077.jsonl.gzData3.8 MBd27896008996
data/step_000078.jsonl.gzData3.9 MB034c3da7e8da
data/step_000079.jsonl.gzData3.7 MB03c82ca490c4
data/step_000080.jsonl.gzData4.8 MB40265200ac1f
data/step_000081.jsonl.gzData3.4 MBb75f0933536a
data/step_000082.jsonl.gzData3.5 MBa9af29a1df9d
data/step_000083.jsonl.gzData3.4 MB037fcb1facf6
data/step_000084.jsonl.gzData4.0 MBc951df3f9ba4
data/step_000085.jsonl.gzData4.1 MB8d9cf283ac72
data/step_000086.jsonl.gzData3.6 MBcb7924b9afe6
data/step_000087.jsonl.gzData3.4 MBba99096bc7c1
data/step_000088.jsonl.gzData4.1 MBcbd26fa05e32
README.mdDocumentation363 B—
.gitattributesRepository2.5 KB—

License and Download

License
apache-2.0
Access
No access gate
Download from Jiahao Zhang

Released by Jiahao Zhang through its official repository on Hugging Face. Read the license.