Skip to page content
SAVRN
Search Plan a project

Dataset

dpo-finnish-interpreted

by Jonas Lindh jonasaise/dpo-finnish-interpreted

This dataset contains Direct Preference Optimization (DPO) pairs translated and culturally adapted into Finnish from an English source dataset (e.g. ultrafeedback or safe-rlhf).

Rows—
Configurations—
Size14.4 MB
Licensemit
AccessPublicly accessible
Monthly Downloads—

Dataset Card

By Jonas Lindh, published under mit, revision 2b64df4a21b0.

Dataset Card for DPO Finnish Interpreted

Dataset Description

This dataset contains Direct Preference Optimization (DPO) pairs translated and culturally adapted into Finnish from an English source dataset (e.g. ultrafeedback or safe-rlhf).

Origin and Methodology

  • Model Used for Synthesis: kimi-k3 (via local endpoint)
  • Translation Strategy: The dataset was generated using a specialized prompt designed to prevent "translationese" and direct literal translation. The kimi-k3 model was instructed to heavily localize the prompt and the chosen response to fit Finnish culture and idioms. Furthermore, it intentionally generated a rejected response that suffers from poor translation, excessive English loan words, or US-centric hallucinations. This provides a high-quality, targeted signal for DPO/RLHF to teach models to favor authentic, culturally grounded Finnish over robotic dictionary translations.

Structure

The dataset contains the following fields: - prompt: The culturally adapted Finnish prompt. - chosen: The high-quality, authentic Finnish response. - rejected: The poorly translated, uncalibrated response. - target_language: The target language of the row. - dataset_source: The source dataset name. - is_safety_focused: Boolean indicating if this is a safety alignment sample.

License

This dataset inherits the license from its source dataset (MIT or Apache 2.0).

Details

Repository
jonasaise/dpo-finnish-interpreted
Publisher
Jonas Lindh
Task category
Not stated by the source
Tags
dpo, rlhf, ultrafeedback
Size category
Not stated by the source
Languages
fi
Revision
2b64df4a21b013480361e29f4dee467f5015f1c8
Last updated
2026-09-25

Files

257 files, 14.4 MB in total.

Data255 files · 14.4 MB
Documentation1 file · 1.6 KB
Repository1 file · 2.5 KB
Every file
FileTypeSizeSHA-256
data/train-1790246573.parquetData60.8 KB15c860bb0509
data/train-1790246827.parquetData54.3 KB598f58acef01
data/train-1790247085.parquetData58.7 KBc1c9859a5abe
data/train-1790247325.parquetData53.8 KBd4f4257b45bc
data/train-1790247601.parquetData55.3 KBd6ed1bca61f6
data/train-1790247851.parquetData55.0 KBbc6889706961
data/train-1790248104.parquetData62.3 KBef5f7f437d84
data/train-1790248363.parquetData55.0 KBb85cc2ddf27f
data/train-1790248618.parquetData59.3 KB900fd55e7687
data/train-1790248847.parquetData57.0 KB69e595a251f5
data/train-1790249098.parquetData59.5 KB1c90556b9e51
data/train-1790249374.parquetData56.8 KBc5db877899b5
data/train-1790249687.parquetData54.8 KBfd23a22c7cf4
data/train-1790249953.parquetData58.0 KB4c0ad06bc767
data/train-1790250214.parquetData55.3 KBf5aa4506f318
data/train-1790250452.parquetData57.5 KBe8aaae445d44
data/train-1790250692.parquetData52.9 KBc0c581787dd5
data/train-1790250971.parquetData56.6 KBa4c6ed94dffc
data/train-1790251247.parquetData56.8 KB516b80435caa
data/train-1790251481.parquetData53.3 KB7dfc299bb607
data/train-1790251731.parquetData58.2 KBf1d40508e576
data/train-1790252012.parquetData62.3 KB63bc3546b1b1
data/train-1790252288.parquetData55.4 KBe392be19b38d
data/train-1790252533.parquetData56.6 KBc896d319055f
data/train-1790252787.parquetData57.6 KBd2aeb60c7334
data/train-1790253052.parquetData55.9 KB0711539f59d2
data/train-1790253279.parquetData52.1 KB21e77a12f363
data/train-1790253523.parquetData55.6 KBdce01549481e
data/train-1790253773.parquetData55.7 KB2c8b10c1824f
data/train-1790254040.parquetData55.9 KB9ae0339b47dc
data/train-1790254259.parquetData52.0 KBfc26b85216d3
data/train-1790254510.parquetData58.9 KB65c4602c2e31
data/train-1790254725.parquetData53.1 KBafce7026798b
data/train-1790254954.parquetData49.9 KB4aef2b5c542f
data/train-1790255180.parquetData51.3 KBd86600ee4f85
data/train-1790255433.parquetData51.5 KB49e965176146
data/train-1790255695.parquetData55.2 KB02dd2e7d8e1f
data/train-1790255977.parquetData58.2 KBbf0fd6905659
data/train-1790256262.parquetData58.5 KBdd65d25fdf51
data/train-1790256512.parquetData51.8 KB628b228b2fed
data/train-1790256759.parquetData57.4 KBa1e5cce670a3
data/train-1790257030.parquetData61.7 KB479955d0f5cb
data/train-1790257304.parquetData62.3 KB95df4695c86c
data/train-1790257574.parquetData61.2 KBadb279915a44
data/train-1790257834.parquetData57.5 KBd8f2e33acb9e
data/train-1790258115.parquetData53.5 KBaa980b6a5c67
data/train-1790258395.parquetData61.1 KB132c1082219b
data/train-1790258648.parquetData55.3 KB58d7593bd91e
data/train-1790258905.parquetData55.6 KB6c5aea590ebb
data/train-1790259128.parquetData55.2 KB697ed2ff71ea
data/train-1790259378.parquetData53.0 KB8b0997a77cb7
data/train-1790259592.parquetData54.2 KB0aa47062fcba
data/train-1790259857.parquetData56.5 KB8d75f89ebfc0
data/train-1790260139.parquetData54.9 KBc2edd403d1a1
data/train-1790260372.parquetData57.9 KBa9320936c18a
data/train-1790260628.parquetData58.6 KBc5b82c816de0
data/train-1790260847.parquetData52.4 KB2181209e014a
data/train-1790261094.parquetData54.9 KB472a611cb674
data/train-1790261355.parquetData56.4 KBb9ac454ee176
data/train-1790261585.parquetData56.0 KB49347a25bec2
data/train-1790261852.parquetData61.7 KBa4889c10b41b
data/train-1790262082.parquetData56.1 KBbc8435114b68
data/train-1790262334.parquetData55.6 KB8bfac652f62a
data/train-1790262553.parquetData47.7 KBa71ed4ae7bc4
data/train-1790262782.parquetData55.9 KB519d4eaa6414
data/train-1790263022.parquetData56.3 KB191fdb8b025c
data/train-1790263267.parquetData54.8 KB35d97bd02f0e
data/train-1790263486.parquetData53.4 KBd865469e592c
data/train-1790263716.parquetData55.0 KB38583aed4f9a
data/train-1790263962.parquetData55.1 KB1e4c9507cf42
data/train-1790264199.parquetData56.1 KB587fdf144997
data/train-1790264435.parquetData54.3 KB6dd44bfb9d7f
data/train-1790264649.parquetData55.6 KB9b47c9786774
data/train-1790264896.parquetData58.8 KB15fd56cff758
data/train-1790265121.parquetData54.7 KBead2107fdd99
data/train-1790265366.parquetData53.6 KB5115abde8871
data/train-1790265601.parquetData57.0 KBc22f7b1e9fd5
data/train-1790265833.parquetData59.0 KBef6d0041768e
data/train-1790266085.parquetData61.8 KB44f165f0e4df
data/train-1790266341.parquetData60.6 KB6a29003852b7
data/train-1790266584.parquetData54.7 KBf196aa67a16f
data/train-1790266821.parquetData59.9 KBad7ebb06fa82
data/train-1790267051.parquetData59.6 KBa73a4d3f03fa
data/train-1790267291.parquetData63.4 KBcce8c53afa2f
data/train-1790267520.parquetData60.1 KB5567e42d462d
data/train-1790267757.parquetData57.2 KB12c6d9530c85
data/train-1790268024.parquetData61.5 KB241309742f43
data/train-1790268276.parquetData57.0 KB6ed68277d6f5
data/train-1790268507.parquetData55.2 KBadd3cb67e557
data/train-1790268759.parquetData58.3 KBcb5e0c44c132
data/train-1790268992.parquetData53.7 KBe59410284982
data/train-1790269238.parquetData59.8 KBce7f7a6d3de2
data/train-1790269485.parquetData61.8 KBc4759dd85883
data/train-1790269695.parquetData54.4 KB7f60c6329c42
data/train-1790269911.parquetData55.8 KB3da633c85796
data/train-1790270131.parquetData53.8 KB26026e208b04
data/train-1790270381.parquetData56.2 KB671a5cae639b
data/train-1790270630.parquetData60.8 KB4803f0e6081e
data/train-1790270847.parquetData53.6 KB163852307ec3
data/train-1790271085.parquetData51.7 KB2a6893e55992
data/train-1790271348.parquetData50.4 KB69ba4719dd0b
data/train-1790271585.parquetData54.0 KB6d1ee34ea066
data/train-1790271817.parquetData54.6 KBddc253c2d24a
data/train-1790272049.parquetData56.6 KB6d0a26b20fe2
data/train-1790272276.parquetData57.8 KBf4874fb4e8ff
data/train-1790272505.parquetData54.3 KB90a9a83c1068
data/train-1790272734.parquetData53.7 KB3d6ab9237e6d
data/train-1790272969.parquetData56.9 KBb3bbc5559a66
data/train-1790273187.parquetData51.1 KB40ccef8e208e
data/train-1790273429.parquetData57.1 KB53a0f42fc86d
data/train-1790273681.parquetData55.7 KBce6cb1cbb1cb
data/train-1790273922.parquetData59.2 KB3ab7cc0d92e8
data/train-1790274157.parquetData55.7 KBbfd6111bc328
data/train-1790274397.parquetData58.3 KB8de85fb2ccd3
data/train-1790274641.parquetData60.3 KB18108163f257
data/train-1790274877.parquetData59.5 KBcff1351e6736
data/train-1790275129.parquetData51.7 KB88c78a876efc
data/train-1790275353.parquetData58.0 KBca445a01cb70
data/train-1790275606.parquetData56.1 KB4f6c7319d09c
data/train-1790275850.parquetData61.5 KB0581008b0e4e
data/train-1790276079.parquetData57.2 KB9a984aa27b22
data/train-1790276305.parquetData56.8 KB14449ab62bc0
data/train-1790276535.parquetData57.2 KBad80948e8773
data/train-1790276780.parquetData64.0 KB2bea6453b68f
data/train-1790277004.parquetData54.5 KBbc1cb76d55ff
data/train-1790277216.parquetData54.8 KB06eb682af6ee
data/train-1790277437.parquetData58.0 KB007f6bcd692a
data/train-1790277705.parquetData65.2 KBf8de922942f0
data/train-1790277940.parquetData57.3 KB0b6b2477fc52
data/train-1790278161.parquetData52.3 KBffde40acf412
data/train-1790278422.parquetData56.8 KBbf6fe5db63ce
data/train-1790278686.parquetData56.4 KB206797c5d1e2
data/train-1790278936.parquetData56.9 KB7ec7f70728f0
data/train-1790279186.parquetData58.3 KB25e6d0fcdecc
data/train-1790279429.parquetData60.4 KBbc35b4c4b9fb
data/train-1790279659.parquetData58.2 KB412892b825fa
data/train-1790279890.parquetData53.6 KBfa823da6ab70
data/train-1790280131.parquetData60.0 KB790e3bbddf0b
data/train-1790280369.parquetData59.5 KB50878c01ff02
data/train-1790280612.parquetData58.7 KBe0a6a8e4921d
data/train-1790280824.parquetData51.9 KB10aa915cd1a3
data/train-1790281075.parquetData58.4 KB4fde8b91bbb0
data/train-1790281320.parquetData56.9 KB9c207c70c7e1
data/train-1790281557.parquetData58.6 KB9bc93a5e5db5
data/train-1790281774.parquetData51.2 KB09b7206107bc
data/train-1790281998.parquetData59.9 KBd7582772232e
data/train-1790282205.parquetData54.7 KB73d003281842
data/train-1790282453.parquetData60.7 KBa498ac235055
data/train-1790282709.parquetData60.1 KB3a5b996a6b02
data/train-1790282954.parquetData61.0 KBca710898df24
data/train-1790283175.parquetData51.1 KB29b824871870
data/train-1790283399.parquetData58.9 KBf8fc16a142d5
data/train-1790283637.parquetData58.4 KBbd28766ea303
data/train-1790283894.parquetData51.8 KB6b7fbb9c28d4
data/train-1790284118.parquetData60.0 KB8f97dcb32e3d
data/train-1790284372.parquetData61.7 KB21f4a7f305b2
data/train-1790284613.parquetData55.3 KBa36b91e2384f
data/train-1790284834.parquetData53.5 KB01f4017c807d
data/train-1790285077.parquetData57.8 KB07d321d2deef
data/train-1790285315.parquetData53.5 KB8dbacb8e8d8c
data/train-1790285556.parquetData55.8 KB0fa5f6278828
data/train-1790285782.parquetData53.7 KB27e7f984be8f
data/train-1790286022.parquetData58.2 KB9f2894911eac
data/train-1790286277.parquetData58.8 KB7f76eef74443
data/train-1790286490.parquetData53.3 KBf7404616141d
data/train-1790286707.parquetData54.1 KB5d827623f36c
data/train-1790286948.parquetData56.4 KB142af29cc09c
data/train-1790287190.parquetData57.1 KB0231c53058ae
data/train-1790287435.parquetData58.1 KBe0d76d0dac16
data/train-1790287656.parquetData55.4 KB5deccb4ba9de
data/train-1790287911.parquetData60.4 KBe510b6d510f2
data/train-1790288158.parquetData56.8 KB97b456c6959a
data/train-1790288349.parquetData52.5 KBe6589e86254f
data/train-1790288622.parquetData57.4 KB3aa53e0043bf
data/train-1790288858.parquetData60.0 KB5aa226de5180
data/train-1790289091.parquetData50.6 KBa3ddf500e8c5
data/train-1790289331.parquetData56.5 KBec101d317f0e
data/train-1790289577.parquetData60.7 KBd30f802e7726
data/train-1790289805.parquetData57.9 KB7a411c856e25
data/train-1790290059.parquetData59.1 KB49f283a2332c
data/train-1790290303.parquetData65.1 KBa751427fd1b8
data/train-1790290556.parquetData61.9 KB63177bd49272
data/train-1790290814.parquetData58.0 KBe7cc82e0f7d8
data/train-1790291039.parquetData58.1 KB00ae8dc70ba0
data/train-1790291271.parquetData54.1 KBfb29d1b8d462
data/train-1790291495.parquetData53.5 KB4cb12424f305
data/train-1790291707.parquetData54.1 KBb199b4a3e070
data/train-1790291956.parquetData52.9 KBfaea729f11dc
data/train-1790292190.parquetData54.6 KB3c0ba49c9375
data/train-1790292432.parquetData56.3 KB60d67f887158
data/train-1790292685.parquetData57.9 KB88414b5ff7fc
data/train-1790292910.parquetData51.4 KBa602ef1b142e
data/train-1790293138.parquetData54.5 KB234645d101a4
data/train-1790293382.parquetData55.3 KBfd6548fdf719
data/train-1790293607.parquetData55.1 KBd5d316c9645d
data/train-1790293842.parquetData58.7 KB7a300ca7b9c6
data/train-1790294093.parquetData55.9 KB7534cd9423cd
data/train-1790294344.parquetData58.1 KB47fe370cb9bd
data/train-1790294599.parquetData61.5 KB10b02ca06c93
data/train-1790294856.parquetData61.7 KBfb21c92087df
data/train-1790295096.parquetData56.5 KBfada094ff3ef
data/train-1790295358.parquetData56.0 KB74af52f354f5
data/train-1790295584.parquetData53.5 KBa45ca5b5cffa
data/train-1790295805.parquetData54.7 KB916e36683784
data/train-1790296037.parquetData58.6 KB2fe1f5df7f47
data/train-1790296259.parquetData53.5 KB62b5fd68c510
data/train-1790296483.parquetData52.7 KB212a5964c1b8
data/train-1790296720.parquetData54.2 KBd0db1d18321e
data/train-1790296961.parquetData57.9 KB05e3be3b3886
data/train-1790297191.parquetData58.6 KB521a2382cd9e
data/train-1790297430.parquetData54.1 KB5dd330ba952c
data/train-1790297661.parquetData60.4 KBabe86406485e
data/train-1790297906.parquetData56.6 KB6ab0690baa85
data/train-1790298122.parquetData54.6 KB7f2b97478897
data/train-1790298355.parquetData55.6 KBc2156323d0c2
data/train-1790298611.parquetData59.2 KBf1f8e73d676c
data/train-1790298856.parquetData56.4 KB5ff513524a03
data/train-1790299091.parquetData57.5 KB378bdcfa62b8
data/train-1790299348.parquetData56.1 KBe226f720f7ca
data/train-1790299582.parquetData59.0 KBa56f58bcb80a
data/train-1790299800.parquetData55.6 KB07733bffb775
data/train-1790300040.parquetData56.5 KB21280cf947bb
data/train-1790300281.parquetData55.6 KB76cfe8e0bed1
data/train-1790300485.parquetData58.0 KBe1b3c19b62c9
data/train-1790300690.parquetData50.4 KB484d12418281
data/train-1790300962.parquetData61.9 KBa42fe7dee7ae
data/train-1790301172.parquetData55.0 KB569008c4af97
data/train-1790301408.parquetData53.3 KB3010e03edf96
data/train-1790301623.parquetData49.2 KBbf13dd0efcd8
data/train-1790301869.parquetData57.7 KB46eb8c112ea5
data/train-1790302118.parquetData59.3 KB0beb2436711d
data/train-1790302336.parquetData56.1 KB23a78813662a
data/train-1790302613.parquetData55.9 KB0c6ba2f907dd
data/train-1790302880.parquetData60.6 KBded0977a05a7
data/train-1790303125.parquetData57.2 KB33f106ecb7c1
data/train-1790303343.parquetData58.3 KB0e5dc1e2e6b1
data/train-1790303575.parquetData50.2 KB54c6309c4c4c
data/train-1790303818.parquetData57.4 KBcd139f642bba
data/train-1790304075.parquetData59.1 KBf4edd2fe20a0
data/train-1790304318.parquetData58.0 KBe8544b829b6f
data/train-1790304582.parquetData53.5 KB48101d406751
data/train-1790304822.parquetData63.2 KBbeec2fca1fcd
data/train-1790305068.parquetData58.3 KBa4fd77f21eea
data/train-1790305307.parquetData53.9 KBd1f18c5857fc
data/train-1790305512.parquetData55.0 KB81e9ea5a5a3c
data/train-1790305747.parquetData56.6 KB37b3fb7f4d4e
data/train-1790306020.parquetData60.9 KB015f7064ac01
data/train-1790306263.parquetData56.9 KB18a0eaefe4ea
data/train-1790306495.parquetData57.1 KB836c4fa4da69
data/train-1790306743.parquetData59.4 KB8a09ef32ea12
data/train-1790306973.parquetData58.3 KBc408befb8e83
data/train-1790307188.parquetData53.2 KB88dd9a9a161c
data/train-1790307396.parquetData51.3 KBa439b736075e
data/train-1790307617.parquetData52.5 KBe396d957e4d9
data/train-1790307859.parquetData55.9 KB2a2e4fc1e0a1
README.mdDocumentation1.6 KB—
.gitattributesRepository2.5 KB—

License and Download

License
mit
Access
No access gate
Download from Jonas Lindh

Released by Jonas Lindh through its official repository on Hugging Face. Read the license.