Skip to page content
SAVRN
Search Plan a project

Dataset

maxrl_dppo_qwen3_4B_maxrl_polaris1000_knk_rollouts

by Fahim Tajwar ftajwar/maxrl_dppo_qwen3_4B_maxrl_polaris1000_knk_rollouts

Training rollouts of one verl run, one parquet per trainer step (data/step-.parquet), rows in the order the trainer consumed them; the rollout.n samples of a prompt are contiguous and share uid.

Rows1,089,536
Configurations1
Size11.0 GB
License—
AccessPublicly accessible
Monthly Downloads436

Dataset Card

Training rollouts of one verl run, one parquet per trainer step (data/step-.parquet), rows in the order the trainer consumed them; the rollout.n samples of a prompt are contiguous and share uid. prompttokenids / responsetokenids are the exact token ids (special tokens kept, no padding) and rolloutlogprobs the sampling-time log-prob of every response token. input / output are decoded with skipspecialtokens=True and are for reading only. score is the raw reward.

Excerpt from the card by Fahim Tajwar.

Structure

default 1,089,536 rows

SplitRowsSize
train1,089,53634.0 GB
inputstringoutputstringgtsstringscorefloat64stepint64accfloat64prompt_token_idslistresponse_token_idslistrollout_log_probslistuidstring

Details

Repository
ftajwar/maxrl_dppo_qwen3_4B_maxrl_polaris1000_knk_rollouts
Publisher
Fahim Tajwar
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
9f8e695f028df4bd294733cf3c52a3312ae376c8
Last updated
2026-10-08

Files

268 files, 11.0 GB in total.

Data266 files · 11.0 GB
Documentation1 file · 597 B
Repository1 file · 2.5 KB
Every file
FileTypeSizeSHA-256
data/step-0001.parquetData27.2 MB8b526bd1c13b
data/step-0002.parquetData27.6 MB938f3f85be62
data/step-0003.parquetData27.0 MBe13190c0073c
data/step-0004.parquetData29.0 MBa2fe369706c5
data/step-0005.parquetData28.1 MB351f56ba49ad
data/step-0006.parquetData26.8 MB7302968462d2
data/step-0007.parquetData28.5 MBd3c4319e81cb
data/step-0008.parquetData28.2 MB31bdae8e2cff
data/step-0009.parquetData28.8 MBf4fb78164e04
data/step-0010.parquetData28.6 MB6f3e5d16a80e
data/step-0011.parquetData28.3 MBbbdf3a7bcbf6
data/step-0012.parquetData27.7 MB5f885114b2e3
data/step-0013.parquetData29.8 MB84c85e721883
data/step-0014.parquetData29.4 MB03ff14468e10
data/step-0015.parquetData30.0 MB6a95f6e4bbc5
data/step-0016.parquetData30.3 MB8167d534c957
data/step-0017.parquetData30.9 MB41f27a266834
data/step-0018.parquetData29.5 MB7bfaf5088f5c
data/step-0019.parquetData30.5 MB9deffc956987
data/step-0020.parquetData31.7 MB199ff0964b25
data/step-0021.parquetData32.2 MB08b787ba9d96
data/step-0022.parquetData32.2 MB046685592745
data/step-0023.parquetData32.1 MBfb92f6036d13
data/step-0024.parquetData32.4 MB8615259b9b4f
data/step-0025.parquetData33.6 MBe5113e8f1128
data/step-0026.parquetData34.7 MB24e60fcb4ec7
data/step-0027.parquetData34.3 MBab0f96dc4e79
data/step-0028.parquetData33.6 MBee2c65734dde
data/step-0029.parquetData36.1 MB6f0d68e0a5ed
data/step-0030.parquetData36.1 MB1ae928e691e3
data/step-0031.parquetData34.5 MB74474e607253
data/step-0032.parquetData37.1 MB7793365fe480
data/step-0033.parquetData36.8 MB4826d15f1a15
data/step-0034.parquetData37.1 MB213612b138e7
data/step-0035.parquetData37.4 MB530676cd1416
data/step-0036.parquetData37.9 MBfa2424968440
data/step-0037.parquetData38.0 MB792e8570e308
data/step-0038.parquetData39.7 MB3abf398a51bc
data/step-0039.parquetData39.1 MB74733117da9b
data/step-0040.parquetData38.5 MB92496d3af000
data/step-0041.parquetData40.0 MB29b0df2131d9
data/step-0042.parquetData41.1 MB5dc4bdc3ee1d
data/step-0043.parquetData40.2 MB39564692e065
data/step-0044.parquetData41.2 MB982659e5c9eb
data/step-0045.parquetData41.2 MB651d8b3bae56
data/step-0046.parquetData39.6 MB9e04ca62efb4
data/step-0047.parquetData39.6 MBe164093f9629
data/step-0048.parquetData40.9 MBfb9a170ed926
data/step-0049.parquetData41.2 MBb017103270eb
data/step-0050.parquetData41.2 MB034f2384752d
data/step-0051.parquetData41.1 MB3c39194c57ed
data/step-0052.parquetData42.1 MB8ef0bae40b6d
data/step-0053.parquetData44.0 MBbd5a9d5f7ef9
data/step-0054.parquetData42.5 MB6b2a9e06c867
data/step-0055.parquetData42.4 MB6eb86e129ff0
data/step-0056.parquetData42.0 MB622d3aca82cb
data/step-0057.parquetData42.6 MBdbf5d4e41e42
data/step-0058.parquetData41.5 MB54bf0f86ae9d
data/step-0059.parquetData41.2 MB871d3a5cced2
data/step-0060.parquetData42.3 MB425185aa0a5a
data/step-0061.parquetData42.9 MB4974e42c5558
data/step-0062.parquetData40.8 MB2a1867a70403
data/step-0063.parquetData42.8 MB86da68893105
data/step-0064.parquetData44.0 MBe863673a0a20
data/step-0065.parquetData43.3 MBe8aec592183c
data/step-0066.parquetData42.5 MBef79b53def7a
data/step-0067.parquetData43.9 MBcaf65897028f
data/step-0068.parquetData43.5 MB00c9f225f157
data/step-0069.parquetData41.5 MB6ad91400eea8
data/step-0070.parquetData44.2 MB5e8c597e17b3
data/step-0071.parquetData43.6 MB998dfd7170e6
data/step-0072.parquetData43.9 MB4ada939343dc
data/step-0073.parquetData43.3 MBe56846a1b565
data/step-0074.parquetData44.3 MB5b44647ba920
data/step-0075.parquetData43.4 MB3c7128b2da99
data/step-0076.parquetData42.7 MB8130de5a7cae
data/step-0077.parquetData42.5 MB22406c68b185
data/step-0078.parquetData42.4 MB13eb50659faa
data/step-0079.parquetData43.5 MBd3e8f0bbb9bf
data/step-0080.parquetData44.2 MBd67a2d8235f2
data/step-0081.parquetData43.1 MB4b0a6f75d188
data/step-0082.parquetData43.7 MBc7bc7cd8ae07
data/step-0083.parquetData44.5 MB24a6dbbba960
data/step-0084.parquetData43.6 MBa2657c1db3c6
data/step-0085.parquetData43.4 MB132292a3be36
data/step-0086.parquetData43.3 MB61066471a637
data/step-0087.parquetData41.8 MB86f9788f6e17
data/step-0088.parquetData42.0 MBaf3573ef8df9
data/step-0089.parquetData44.9 MB849464442a19
data/step-0090.parquetData44.4 MBf0d842e0013a
data/step-0091.parquetData43.2 MBcbd965094baa
data/step-0092.parquetData43.6 MB62bf71317089
data/step-0093.parquetData43.7 MBd8be05b17b82
data/step-0094.parquetData43.8 MBfeb9e2f00e88
data/step-0095.parquetData45.2 MB7068d83a8e03
data/step-0096.parquetData44.3 MB2ddeb1a1f307
data/step-0097.parquetData45.8 MB366314bee982
data/step-0098.parquetData42.6 MBf266c62061ea
data/step-0099.parquetData42.3 MB29a6c24e9553
data/step-0100.parquetData43.1 MB98d16cec59b6
data/step-0101.parquetData43.8 MB2868459ebcbb
data/step-0102.parquetData43.1 MB8debca31e0d1
data/step-0103.parquetData42.6 MB0389c0913405
data/step-0104.parquetData43.6 MB7e3d4714b52a
data/step-0105.parquetData43.9 MB16db12a1ae1f
data/step-0106.parquetData43.4 MB1e96f0792df8
data/step-0107.parquetData43.6 MBa41b31c65474
data/step-0108.parquetData43.6 MB4e9fd5ad81f5
data/step-0109.parquetData45.2 MB26eded8c3235
data/step-0110.parquetData43.0 MBd1d2f5567c24
data/step-0111.parquetData43.2 MB5823e66aae66
data/step-0112.parquetData45.8 MB6a70ce90bf99
data/step-0113.parquetData43.5 MB5b5678926981
data/step-0114.parquetData43.7 MB0af947c5991a
data/step-0115.parquetData44.3 MB4c4f583e841b
data/step-0116.parquetData44.3 MB56dfcfc2c06a
data/step-0117.parquetData45.3 MBdad557ba5bc7
data/step-0118.parquetData43.6 MBb5dbe7488403
data/step-0119.parquetData44.4 MB565ce10105b4
data/step-0120.parquetData42.9 MB85e9ebff31e9
data/step-0121.parquetData45.3 MB9e00b82d4add
data/step-0122.parquetData44.1 MB8f84eb6d1900
data/step-0123.parquetData44.3 MB96501da98283
data/step-0124.parquetData44.2 MB26a8d0f02db9
data/step-0125.parquetData43.5 MB5ca4aae99c2e
data/step-0126.parquetData43.2 MB5dabf35bd3b1
data/step-0127.parquetData43.5 MBa9e8cf70c154
data/step-0128.parquetData43.3 MBfcc599d07222
data/step-0129.parquetData44.1 MB992142e7cd50
data/step-0130.parquetData42.4 MB49033ec570cc
data/step-0131.parquetData43.8 MB7d1694e1f7ca
data/step-0132.parquetData45.8 MB9dff46d5cf5f
data/step-0133.parquetData43.6 MB4093ea0e5707
data/step-0134.parquetData42.6 MB65509cf42dc7
data/step-0135.parquetData45.1 MBebb673a0b0fe
data/step-0136.parquetData44.2 MB3fcf3d4c0d81
data/step-0137.parquetData43.3 MBfbd4ade1fceb
data/step-0138.parquetData45.1 MB5d5bc1dcbe22
data/step-0139.parquetData44.4 MB2e190b6736a3
data/step-0140.parquetData43.5 MB605c9dc5b482
data/step-0141.parquetData44.6 MB359f6b0a5359
data/step-0142.parquetData43.0 MBbeef8788cfa1
data/step-0143.parquetData42.4 MBf875baa8b365
data/step-0144.parquetData41.6 MB5dadbc2640d2
data/step-0145.parquetData41.8 MB5e6ee0914c63
data/step-0146.parquetData44.1 MB64cc499ea38e
data/step-0147.parquetData42.0 MB17713eff9904
data/step-0148.parquetData43.2 MB30860d0776ee
data/step-0149.parquetData44.8 MBa99df74818b6
data/step-0150.parquetData44.5 MBa9c27b45547b
data/step-0151.parquetData44.3 MBa3cfbdcdd7a6
data/step-0152.parquetData44.4 MBff283a1b59a7
data/step-0153.parquetData43.2 MB2c3809a829be
data/step-0154.parquetData45.1 MBac9780544e2e
data/step-0155.parquetData42.5 MB50efda8f84e1
data/step-0156.parquetData42.1 MB1099d2421e6f
data/step-0157.parquetData43.7 MB98b1b044ec7a
data/step-0158.parquetData44.1 MB5496ee852142
data/step-0159.parquetData44.7 MBc252f0635225
data/step-0160.parquetData41.2 MBb24858952d90
data/step-0161.parquetData45.1 MB9e96dfb50e05
data/step-0162.parquetData44.1 MB6fe24ee174fb
data/step-0163.parquetData43.3 MBe41a68f8f4e6
data/step-0164.parquetData44.8 MB71d8979f5973
data/step-0165.parquetData43.8 MB2a1ec7a66124
data/step-0166.parquetData44.8 MB6169581867e7
data/step-0167.parquetData45.9 MB366df3086f5d
data/step-0168.parquetData45.5 MB38e12a665303
data/step-0169.parquetData45.0 MBde944e587a1e
data/step-0170.parquetData43.0 MB1685e95a7628
data/step-0171.parquetData45.7 MB5024d61d06fd
data/step-0172.parquetData44.7 MBb6be7b5d2d05
data/step-0173.parquetData42.7 MBb90cf2514bb3
data/step-0174.parquetData45.3 MBd7556878b6d7
data/step-0175.parquetData44.0 MBa558e1e927f3
data/step-0176.parquetData44.0 MBe0d5b2aca02f
data/step-0177.parquetData43.8 MB535baa83ac3d
data/step-0178.parquetData43.5 MB71b436fc4dd5
data/step-0179.parquetData42.8 MB28b8affb09ef
data/step-0180.parquetData41.9 MB276a16ce69e2
data/step-0181.parquetData42.8 MBac2edd645eb9
data/step-0182.parquetData44.5 MBbc735d92005a
data/step-0183.parquetData42.5 MBb7adc689dcbb
data/step-0184.parquetData42.6 MBe478f23abf3c
data/step-0185.parquetData41.0 MBef95ec109810
data/step-0186.parquetData42.4 MBb3954c9c1219
data/step-0187.parquetData42.5 MBcb671eb0bdb0
data/step-0188.parquetData42.2 MBecb8265e454a
data/step-0189.parquetData43.7 MB73595635a99a
data/step-0190.parquetData41.8 MB1dfcdc59e454
data/step-0191.parquetData42.2 MB0572579754a0
data/step-0192.parquetData41.7 MBdfb769231981
data/step-0193.parquetData41.8 MBea619105532d
data/step-0194.parquetData41.8 MB626802f90d5e
data/step-0195.parquetData43.2 MB05943b326fb5
data/step-0196.parquetData43.1 MB27b78c16d795
data/step-0197.parquetData43.8 MB8a85cb60f810
data/step-0198.parquetData41.8 MB1b992c7ddb4a
data/step-0199.parquetData44.2 MB4ed3f72ba882
data/step-0200.parquetData42.3 MB48153da0db25
data/step-0201.parquetData43.0 MBe0db11f7dd31
data/step-0202.parquetData43.2 MBd5ab4ab53f9e
data/step-0203.parquetData41.8 MB0e9ebbd8f0cc
data/step-0204.parquetData43.3 MB5ecbda3f8d1b
data/step-0205.parquetData42.1 MB6e9d81cd397e
data/step-0206.parquetData42.4 MBf4807b5eda37
data/step-0207.parquetData42.5 MB8f06af378d79
data/step-0208.parquetData41.4 MB1f6c03d858eb
data/step-0209.parquetData42.1 MB9d2398374a0a
data/step-0210.parquetData41.9 MB5397cc9a349d
data/step-0211.parquetData42.0 MB944fa0da6600
data/step-0212.parquetData40.9 MBbde28ae9eecd
data/step-0213.parquetData43.1 MB772ef27e769e
data/step-0214.parquetData42.3 MB3e1e3e3888e4
data/step-0215.parquetData43.2 MBf6fd38fab32f
data/step-0216.parquetData41.6 MB1a9db873e270
data/step-0217.parquetData42.4 MB6c6df620490d
data/step-0218.parquetData41.6 MBf71d1e971631
data/step-0219.parquetData41.0 MBf631782cc172
data/step-0220.parquetData41.4 MBa236dd0a50b5
data/step-0221.parquetData41.5 MB0538bdedede2
data/step-0222.parquetData41.5 MBc0599e175443
data/step-0223.parquetData41.2 MB9f386fbbcceb
data/step-0224.parquetData41.3 MB92ebd0e1919b
data/step-0225.parquetData42.0 MBfba95e3f271e
data/step-0226.parquetData42.3 MB7aa17966a6a3
data/step-0227.parquetData41.3 MB64ded7e9d4f4
data/step-0228.parquetData40.0 MBfaba13925aa2
data/step-0229.parquetData40.5 MB9195a045297b
data/step-0230.parquetData40.7 MB8c33267804ba
data/step-0231.parquetData38.7 MBd15aa2cf0f34
data/step-0232.parquetData39.8 MB000cf7cd19dc
data/step-0233.parquetData41.7 MBe3d84a13654b
data/step-0234.parquetData41.3 MB9d920ec59090
data/step-0235.parquetData40.2 MBf03c44531e33
data/step-0236.parquetData39.9 MB1ddd624f9fdb
data/step-0237.parquetData40.1 MB345fd71dde31
data/step-0238.parquetData40.4 MB6acd9afd10f0
data/step-0239.parquetData40.1 MBfdd041ca8e6e
data/step-0240.parquetData39.2 MB3fc140d1b642
data/step-0241.parquetData39.1 MBf15d0e897e89
data/step-0242.parquetData40.3 MB119291791951
data/step-0243.parquetData38.8 MBcfca77b07fce
data/step-0244.parquetData39.5 MB69c111b4e9ec
data/step-0245.parquetData41.6 MBfd1dab50f429
data/step-0246.parquetData41.2 MB05be90a563ae
data/step-0247.parquetData39.6 MB4d3fef87d8cc
data/step-0248.parquetData41.1 MB4c6fbd287639
data/step-0249.parquetData42.1 MB883ed1d76aa3
data/step-0250.parquetData41.3 MB609283e9e228
data/step-0251.parquetData42.0 MBbb92d0227320
data/step-0252.parquetData43.4 MB4c3693baae1f
data/step-0253.parquetData40.5 MB7f6fbae10048
data/step-0254.parquetData42.5 MB70efed0a3f29
data/step-0255.parquetData44.3 MB6639357092f6
data/step-0256.parquetData43.0 MB515b04ddfe8e
data/step-0257.parquetData43.4 MBf4239bd91b7b
data/step-0258.parquetData44.3 MB2c24e8f5357b
data/step-0259.parquetData43.1 MB972d55fc9122
data/step-0260.parquetData44.2 MBeba7656e0331
data/step-0261.parquetData43.4 MBc496ae49b936
data/step-0262.parquetData42.8 MB3dd3c8f4ce2c
data/step-0263.parquetData43.1 MB623705fe2c99
data/step-0264.parquetData42.4 MB381356402182
data/step-0265.parquetData43.7 MB7dbb970e0dc9
data/step-0266.parquetData43.0 MBe852ce3e830c
README.mdDocumentation597 B—
.gitattributesRepository2.5 KB—

License and Download

License
Not stated by the source
Access
No access gate
Download from Fahim Tajwar

Released by Fahim Tajwar through its official repository on Hugging Face.