student-multilang-XLMR-14jun

slimaneMakh/student-multilang-XLMR-14jun

student-multilang-XLMR-14jun is a sentence similarity model from slimaneMakh. Use it when you need a score for how close two texts are. It is set up for sentence-transformers.

Downloads · 30 days

19% of all-time downloads

All-time downloads

484

Public

Parameters

278M

1.1 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors1.1 GB · 98%

At a glance

Task: Sentence Similarity
Library: sentence-transformers
Model type: xlm-roberta
Access: Public
Created: Jun 14, 2024
Updated: Jun 14, 2024
SHA: a3d4df3e

Base models

FacebookAI/xlm-roberta-base

Model card

From the Hugging Face model README

SentenceTransformer based on FacebookAI/xlm-roberta-base

This is a sentence-transformers model finetuned from FacebookAI/xlm-roberta-base on the en-fr, en-fi, en-pl, en-sv, en-de, en-it, en-pt, en-no, en-nb, en-de-de, en-es, en-cs, en-nl, en-da, en-lt, en-is, en-sl, en-sv-se, en-fi-fi, en-en-gb, en-lv, en-el and en-et datasets. It maps sentences & paragraphs to a 768-dimensional dense vector space and can be used for semantic textual similarity, semantic search, paraphrase mining, text classification, clustering, and more.

Model Details

Model Description

Model Type: Sentence Transformer
Base model: FacebookAI/xlm-roberta-base
Maximum Sequence Length: 128 tokens
Output Dimensionality: 768 tokens
Similarity Function: Cosine Similarity
Training Datasets:
- en-fr
- en-fi
- en-pl
- en-sv
- en-de
- en-it
- en-pt
- en-no
- en-nb
- en-de-de
- en-es
- en-cs
- en-nl
- en-da
- en-lt
- en-is
- en-sl
- en-sv-se
- en-fi-fi
- en-en-gb
- en-lv
- en-el
- en-et

Model Sources

Documentation: Sentence Transformers Documentation
Repository: Sentence Transformers on GitHub
Hugging Face: Sentence Transformers on Hugging Face

Full Model Architecture

SentenceTransformer(
  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: XLMRobertaModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False, 'pooling_mode_weightedmean_tokens': False, 'pooling_mode_lasttoken': False, 'include_prompt': True})
)

Usage

Direct Usage (Sentence Transformers)

First install the Sentence Transformers library:

pip install -U sentence-transformers

Then you can load this model and run inference.

from sentence_transformers import SentenceTransformer

# Download from the 🤗 Hub
model = SentenceTransformer("slimaneMakh/student-multilang-XLMR-14jun")
# Run inference
sentences = [
    'Financial asset investments',
    'Financne nalozbe',
    'activities',
]
embeddings = model.encode(sentences)
print(embeddings.shape)
# [3, 768]

# Get the similarity scores for the embeddings
similarities = model.similarity(embeddings, embeddings)
print(similarities.shape)
# [3, 3]

Evaluation

Metrics

Knowledge Distillation

Dataset: en-fr
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-18.7979

Translation

Dataset: en-fr
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0026
trg2src_accuracy	0.0022
mean_accuracy	0.0024

Knowledge Distillation

Dataset: en-fi
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.079

Translation

Dataset: en-fi
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0055
trg2src_accuracy	0.0048
mean_accuracy	0.0052

Knowledge Distillation

Dataset: en-pl
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-18.9324

Translation

Dataset: en-pl
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0031
trg2src_accuracy	0.0027
mean_accuracy	0.0029

Knowledge Distillation

Dataset: en-sv
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.0325

Translation

Dataset: en-sv
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0037
trg2src_accuracy	0.004
mean_accuracy	0.0038

Knowledge Distillation

Dataset: en-de
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.2001

Translation

Dataset: en-de
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0026
trg2src_accuracy	0.0027
mean_accuracy	0.0027

Knowledge Distillation

Dataset: en-it
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.0771

Translation

Dataset: en-it
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0035
trg2src_accuracy	0.0036
mean_accuracy	0.0036

Knowledge Distillation

Dataset: en-pt
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.0009

Translation

Dataset: en-pt
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0084
trg2src_accuracy	0.0081
mean_accuracy	0.0083

Knowledge Distillation

Dataset: en-no
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-20.6052

Translation

Dataset: en-no
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.011
trg2src_accuracy	0.0123
mean_accuracy	0.0117

Knowledge Distillation

Dataset: en-nb
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-20.6013

Translation

Dataset: en-nb
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0127
trg2src_accuracy	0.0127
mean_accuracy	0.0127

Knowledge Distillation

Dataset: en-de-de
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-20.8682

Translation

Dataset: en-de-de
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0282
trg2src_accuracy	0.0282
mean_accuracy	0.0282

Knowledge Distillation

Dataset: en-es
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-18.8438

Translation

Dataset: en-es
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0051
trg2src_accuracy	0.0047
mean_accuracy	0.0049

Knowledge Distillation

Dataset: en-cs
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.1286

Translation

Dataset: en-cs
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0112
trg2src_accuracy	0.0145
mean_accuracy	0.0129

Knowledge Distillation

Dataset: en-nl
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.8483

Translation

Dataset: en-nl
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.007
trg2src_accuracy	0.0082
mean_accuracy	0.0076

Knowledge Distillation

Dataset: en-da
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.3856

Translation

Dataset: en-da
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0116
trg2src_accuracy	0.0126
mean_accuracy	0.0121

Knowledge Distillation

Dataset: en-lt
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-20.485

Translation

Dataset: en-lt
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0109
trg2src_accuracy	0.0109
mean_accuracy	0.0109

Knowledge Distillation

Dataset: en-is
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.2169

Translation

Dataset: en-is
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0072
trg2src_accuracy	0.0093
mean_accuracy	0.0083

Knowledge Distillation

Dataset: en-sl
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-18.1531

Translation

Dataset: en-sl
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0112
trg2src_accuracy	0.014
mean_accuracy	0.0126

Knowledge Distillation

Dataset: en-sv-se
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-17.6476

Translation

Dataset: en-sv-se
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0234
trg2src_accuracy	0.0208
mean_accuracy	0.0221

Knowledge Distillation

Dataset: en-fi-fi
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-19.282

Translation

Dataset: en-fi-fi
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.018
trg2src_accuracy	0.018
mean_accuracy	0.018

Knowledge Distillation

Dataset: en-en-gb
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-23.5088

Translation

Dataset: en-en-gb
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0126
trg2src_accuracy	0.0167
mean_accuracy	0.0146

Knowledge Distillation

Dataset: en-lv
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-18.0377

Translation

Dataset: en-lv
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0048
trg2src_accuracy	0.0095
mean_accuracy	0.0071

Knowledge Distillation

Dataset: en-el
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-23.5207

Translation

Dataset: en-el
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0513
trg2src_accuracy	0.0513
mean_accuracy	0.0513

Knowledge Distillation

Dataset: en-et
Evaluated with <code>MSEEvaluator</code>

Metric	Value
negative_mse	-17.5146

Translation

Dataset: en-et
Evaluated with <code>TranslationEvaluator</code>

Metric	Value
src2trg_accuracy	0.0192
trg2src_accuracy	0.0192
mean_accuracy	0.0192

Training Details

Training Datasets

en-fr

Dataset: en-fr
Size: 63,449 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 10.09 tokens</li><li>max: 30 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.1 tokens</li><li>max: 24 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.0459553524851799, 0.36456549167633057, 0.36365264654159546, 0.6452828645706177, -0.4019026756286621, ...]</code>	<code>Net income for the period attributable to shareholders</code>	<code>Resultat de lexercice</code>
<code>[0.44971197843551636, 0.9621334075927734, -0.0879441499710083, -0.08917804807424545, 0.002839124295860529, ...]</code>	<code>Podatek dochodowy</code>	<code>Impots</code>
<code>[0.3880807161331177, 0.19511738419532776, -0.13357722759246826, 0.25993096828460693, 0.0716109424829483, ...]</code>	<code>AttributabletotheshareholdersofKvikabankihf</code>	<code>aux actionnaires de la Societe</code>

Loss: <code>MSELoss</code>

en-fi

Dataset: en-fi
Size: 18,428 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.67 tokens</li><li>max: 31 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 7.69 tokens</li><li>max: 17 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.24573877453804016, 0.5694760680198669, 0.45771917700767517, -0.13942377269268036, -0.2597014904022217, ...]</code>	<code>Shareholders of Copenhagen Airports AS</code>	<code>Emoyhtion osakkeenomistajille</code>
<code>[0.5077632665634155, 0.8774086236953735, -0.3499397933483124, -0.6389203667640686, 0.026370976120233536, ...]</code>	<code>Income tax benefit expense</code>	<code>Income taxes</code>
<code>[0.9414718747138977, -0.24161840975284576, 0.41289815306663513, 0.10003143548965454, -1.092337965965271, ...]</code>	<code>Result</code>	<code>Emoyrityksen osakkeenomistajille kuuluvasta tuloksesta laskettu</code>

Loss: <code>MSELoss</code>

en-pl

Dataset: en-pl
Size: 45,054 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 10.09 tokens</li><li>max: 29 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 12.42 tokens</li><li>max: 39 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.09482160955667496, 0.7886450886726379, 0.23035818338394165, 0.21230120956897736, 0.33353161811828613, ...]</code>	<code>Changes in deferred taxes directly recognized in other comprehensive income</code>	<code>Podatek dochodowy dotyczacy innych calkowitych dochodow</code>
<code>[-0.15856720507144928, 0.6147034168243408, -0.25085723400115967, -0.5494844913482666, -0.526219367980957, ...]</code>	<code>Diluted from continuing operations</code>	<code>Rozwodniony zysk strata na jedna akcje</code>
<code>[-0.1696387380361557, -0.23339493572711945, -0.7045446038246155, -0.3721548914909363, -0.36909934878349304, ...]</code>	<code>CASH FLOW RESULTING FROM OPERATING ACTIVITIES</code>	<code>Srodki pieniezne netto z dzialalnosci operacyjnej</code>

Loss: <code>MSELoss</code>

en-sv

Dataset: en-sv
Size: 37,354 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.67 tokens</li><li>max: 36 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.34 tokens</li><li>max: 21 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.2742433547973633, -0.4345971345901489, -0.28529638051986694, -0.06954757869243622, -1.822569489479065, ...]</code>	<code>grupe</code>	<code>moderbolagets aktieagare</code>
<code>[0.04750566929578781, 0.2545453608036041, 0.3464582860469818, 0.22448834776878357, -0.0583755262196064, ...]</code>	<code>Total comprehensive income for the year attributable to owners of the parent Company</code>	<code>Moderbolagets aktieagare</code>
<code>[0.045431576669216156, 0.3078455924987793, -0.06083355098962784, -0.5454118847846985, 0.5727013349533081, ...]</code>	<code>Repayment of obligations under lease arrangements</code>	<code>Amortering av skuld</code>

Loss: <code>MSELoss</code>

en-de

Dataset: en-de
Size: 45,253 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.92 tokens</li><li>max: 31 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.18 tokens</li><li>max: 23 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.086859792470932, 0.7745860815048218, -0.08605925738811493, 0.37508440017700195, -0.9738988876342773, ...]</code>	<code>adjustments of investments in subsidiaries</code>	<code>Wahrungsumrechnungsdifferenzen</code>
<code>[-0.05315065383911133, 0.0072781918570399284, -0.2516656517982483, -0.4747457504272461, -1.1008282899856567, ...]</code>	<code>LOSS FROM CONTINUING OPERATIONS</code>	<code>Ergebnis nach Ertragsteuern</code>
<code>[0.14867287874221802, 1.0406593084335327, -0.17914682626724243, -0.6161922812461853, 0.14850790798664093, ...]</code>	<code>Taxation paid received</code>	<code>Ertragsteueraufwand ertrag</code>

Loss: <code>MSELoss</code>

en-it

Dataset: en-it
Size: 34,682 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.98 tokens</li><li>max: 26 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.39 tokens</li><li>max: 20 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.5695832371711731, 0.02826128527522087, 0.1920386552810669, 0.40783414244651794, -1.2495031356811523, ...]</code>	<code>Current financial receivables</code>	<code>Titoli in portafoglio</code>
<code>[0.662227988243103, 0.6725629568099976, 0.22833657264709473, 0.054810211062431335, -0.40215858817100525, ...]</code>	<code>Proceeds from sale of assets</code>	<code>Attivita destinate alla vendita</code>
<code>[0.1357184797525406, 0.7814697623252869, 0.3390173614025116, -0.10204766690731049, -0.3055779039859772, ...]</code>	<code>Profit before income tax</code>	<code>Risultato netto</code>

Loss: <code>MSELoss</code>

en-pt

Dataset: en-pt
Size: 7,300 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.84 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 6.71 tokens</li><li>max: 12 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.008626206777989864, 0.6093286275863647, 0.08171450346708298, 1.162959337234497, 0.6411553025245667, ...]</code>	<code>Interest received by the Barclays Bank Group was m</code>	<code>Juros recebidos</code>
<code>[-0.27057403326034546, 0.2500847578048706, -0.07353457063436508, 0.5000247955322266, -0.07040926814079285, ...]</code>	<code>Other liabilities</code>	<code>Outros passivos</code>
<code>[-0.03809820115566254, 0.1842460036277771, -0.08849599212408066, -0.844947338104248, 0.7437804341316223, ...]</code>	<code>Payment of obligations under leases</code>	<code>Passivos de locacao</code>

Loss: <code>MSELoss</code>

en-no

Dataset: en-no
Size: 3,602 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.04 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 5.4 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.19592446088790894, 0.5323967337608337, 0.21345381438732147, -0.4241628348827362, -0.0008733272552490234, ...]</code>	<code>of the parent company</code>	<code>income</code>
<code>[-0.05730602145195007, 0.16925856471061707, -0.16081246733665466, -1.6013731956481934, 0.6432715654373169, ...]</code>	<code>Employee charges and benefits expenses</code>	<code>Personalkostnader</code>
<code>[0.053435444831848145, -0.08411762863397598, 0.7841566801071167, 0.822182834148407, -0.3946605324745178, ...]</code>	<code>in expected credit losses net</code>	<code>totalresultat</code>

Loss: <code>MSELoss</code>

en-nb

Dataset: en-nb
Size: 3,446 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.39 tokens</li><li>max: 34 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 5.83 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.6152929663658142, 1.0328565835952759, -0.48867374658584595, 0.6196318864822388, -1.0412869453430176, ...]</code>	<code>Note b</code>	<code>Andre driftskostnader</code>
<code>[-0.08955559879541397, 0.07031169533729553, -0.4530458450317383, 0.6429653763771057, -0.17220227420330048, ...]</code>	<code>Profitloss for the period</code>	<code>Resultat</code>
<code>[-0.2092481404542923, 0.8907342553138733, -0.2213028073310852, 0.19046330451965332, 0.36781418323516846, ...]</code>	<code>Tax on profitloss</code>	<code>Skattekostnad</code>

Loss: <code>MSELoss</code>

en-de-de

Dataset: en-de-de
Size: 623 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.53 tokens</li><li>max: 31 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 9.83 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.15285512804985046, 0.24292221665382385, -0.21986141800880432, -0.12183597683906555, -0.8729998469352722, ...]</code>	<code>Ikkekontrollerende eierinteresse</code>	<code>davon den nicht beherrschenden Anteilen zuzurechnen</code>
<code>[0.7105820178985596, 0.6940978765487671, 0.29005366563796997, 0.33401334285736084, 0.05582822486758232, ...]</code>	<code>Total net revenue</code>	<code>Umsatzerlose</code>
<code>[-0.20316101610660553, 0.9045584797859192, -0.2203243523836136, -1.074849247932434, -0.4881342351436615, ...]</code>	<code>Caixa e equivalentes de caixa</code>	<code>Zahlungsmittel und Zahlungsmittelaquivalente</code>

Loss: <code>MSELoss</code>

en-es

Dataset: en-es
Size: 28,719 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.47 tokens</li><li>max: 28 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.48 tokens</li><li>max: 23 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.172838494181633, 0.43473777174949646, 0.3958137333393097, 0.1424863040447235, -0.8349866271018982, ...]</code>	<code>Increase in trade receivables and other assets</code>	<code>Clientes y otras cuentas a cobrar</code>
<code>[0.5418481826782227, 0.5917099714279175, 0.1668325960636139, 0.3066450357437134, -1.260878324508667, ...]</code>	<code>Increase in trade and other receivables and advances paid</code>	<code>Clientes y otras cuentas a cobrar</code>
<code>[-0.2715812921524048, 0.05829544737935066, -0.4542696177959442, -0.029009468853473663, -0.7529364824295044, ...]</code>	<code>Total Comprehensive Loss for the year wholly attributable to Equity Holders of the Parent Company</code>	<code>Atribuible a la sociedad dominante</code>

Loss: <code>MSELoss</code>

en-cs

Dataset: en-cs
Size: 2,203 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 10.1 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 7.9 tokens</li><li>max: 23 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.06290890276432037, 0.5762706398963928, -0.024871770292520523, 0.22431252896785736, -0.6742631196975708, ...]</code>	<code>Udzialy niekontrolujace</code>	<code>Nekontrolnim podilum</code>
<code>[0.39093080163002014, -0.009997962974011898, 0.24490250647068024, 0.9013416171073914, -0.796424388885498, ...]</code>	<code>Profit for the year attributable to ordinary Shareholders</code>	<code>Akcionarum materske spolecnosti</code>
<code>[-0.23978163301944733, 0.484517902135849, -0.3151543438434601, 0.1443774700164795, -0.16455821692943573, ...]</code>	<code>Avsetning for forpliktelser</code>	<code>Rezervy</code>

Loss: <code>MSELoss</code>

en-nl

Dataset: en-nl
Size: 8,101 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.39 tokens</li><li>max: 34 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 7.01 tokens</li><li>max: 12 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.43313074111938477, -0.23663929104804993, -0.0008638567524030805, 0.21914006769657135, -1.1042245626449585, ...]</code>	<code>Shareholders of FGC UES</code>	<code>Aandeelhouders van de moedermaatschappij</code>
<code>[0.5194972157478333, 0.45368078351020813, 0.5302746295928955, 0.2755521535873413, -0.3021118640899658, ...]</code>	<code>Noncontrolling interest</code>	<code>Belang van derden</code>
<code>[0.9302910566329956, 0.7344815731048584, 0.6589862108230591, 0.1774829477071762, 0.528937578201294, ...]</code>	<code>Debt</code>	<code>Leningen</code>

Loss: <code>MSELoss</code>

en-da

Dataset: en-da
Size: 4,554 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.91 tokens</li><li>max: 31 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 7.54 tokens</li><li>max: 26 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.016798147931694984, 0.7280638813972473, 0.1259734034538269, -0.07660696655511856, -0.20033679902553558, ...]</code>	<code>Provisions current portion</code>	<code>Hensatte forpligtelser</code>
<code>[-0.07381738722324371, -0.07786396145820618, -0.21328210830688477, 0.18608279526233673, -0.3095148205757141, ...]</code>	<code>or loss</code>	<code>Kursreguleringer</code>
<code>[-0.4245157241821289, 0.4695541262626648, 0.05997037887573242, 0.2986871004104614, 0.011750679463148117, ...]</code>	<code>assets depreciation</code>	<code>Af og nedskrivninger</code>

Loss: <code>MSELoss</code>

en-lt

Dataset: en-lt
Size: 2,998 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.39 tokens</li><li>max: 31 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 8.65 tokens</li><li>max: 12 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.2119722217321396, 0.5226094722747803, -0.3225395679473877, 0.6458964347839355, -0.22873802483081818, ...]</code>	<code>NOTE</code>	<code>Atsargos</code>
<code>[0.5478602647781372, 0.3326689302921295, -0.14589856564998627, 0.5814526677131653, 0.5692975521087646, ...]</code>	<code>Repayment of loan</code>	<code>Paskolu grazinimas</code>
<code>[0.2744126319885254, 0.5255246162414551, 0.05724802985787392, 0.25815054774284363, -0.766740620136261, ...]</code>	<code>Attributable to the owners of the Company</code>	<code>Bendroves akcininkams</code>

Loss: <code>MSELoss</code>

en-is

Dataset: en-is
Size: 2,138 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.65 tokens</li><li>max: 29 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 10.54 tokens</li><li>max: 15 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.037829890847206116, 1.1669130325317383, 0.2974126636981964, 0.16161930561065674, 0.022792719304561615, ...]</code>	<code>Tax expenses</code>	<code>Tekjuskattur</code>
<code>[0.11290981620550156, 0.3291318714618683, -0.6060066819190979, 0.029671549797058105, -0.4738736152648926, ...]</code>	<code>Share of profit from Hyundai Glovis</code>	<code>Ahrif hlutdeildarfelaga</code>
<code>[-0.1636863499879837, -0.4239570200443268, 0.2055961787700653, -1.1946961879730225, 0.13549365103244781, ...]</code>	<code>Changes in working capital requirements</code>	<code>Veltufe fra rekstri</code>

Loss: <code>MSELoss</code>

en-sl

Dataset: en-sl
Size: 834 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.04 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 5.48 tokens</li><li>max: 9 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.020984871312975883, -0.31524133682250977, 0.10546927899122238, 1.0089449882507324, -0.592142641544342, ...]</code>	<code>Net cash flows tofrom investing activities</code>	<code>activities</code>
<code>[0.1349133551120758, -0.2043939232826233, 0.2521047592163086, -0.04384709894657135, -0.5578309893608093, ...]</code>	<code>Net cash ows from investing activities</code>	<code>activities</code>
<code>[-0.16783905029296875, 1.331608533859253, 0.9504968523979187, 0.402763694524765, -0.8187195658683777, ...]</code>	<code>Foreign currency translations</code>	<code>Prevedbena rezerva</code>

Loss: <code>MSELoss</code>

en-sv-se

Dataset: en-sv-se
Size: 847 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.93 tokens</li><li>max: 33 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 7.15 tokens</li><li>max: 11 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.14358974993228912, 0.12112939357757568, 0.152898907661438, 0.2965115010738373, -0.6465349197387695, ...]</code>	<code>Cash flow from investing activities</code>	<code>Kassaflode fran investeringsverksamheten</code>
<code>[-0.3012215495109558, -0.6284143924713135, 0.952661395072937, 0.6150138974189758, 1.3908427953720093, ...]</code>	<code>reporting year</code>	<code>Likvida medel</code>
<code>[0.7741854190826416, 0.9692693948745728, -0.48180654644966125, -0.3358636796474457, -1.0314745903015137, ...]</code>	<code>Note c</code>	<code>Personalkostnader</code>

Loss: <code>MSELoss</code>

en-fi-fi

Dataset: en-fi-fi
Size: 874 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.97 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 6 tokens</li><li>mean: 8.69 tokens</li><li>max: 10 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.0959925726056099, -0.0646059587597847, -0.5595968961715698, 0.40048298239707947, -0.0345945879817009, ...]</code>	<code>Soci della controllante</code>	<code>Emoyhtion osakkeenomistajille</code>
<code>[0.07576075196266174, 0.13357341289520264, 0.2546372711658478, 0.0818142369389534, -0.08272691816091537, ...]</code>	<code>ordinary shareholders of the parent company</code>	<code>Emoyhtion osakkeenomistajille</code>
<code>[-0.1580277979373932, 0.6337043642997742, 0.21239566802978516, 0.5370602011680603, -1.064493179321289, ...]</code>	<code>Net gains losses on investments in foreign operations</code>	<code>Muuntoerot</code>

Loss: <code>MSELoss</code>

en-en-gb

Dataset: en-en-gb
Size: 551 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.86 tokens</li><li>max: 38 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 6.33 tokens</li><li>max: 8 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.18707695603370667, 0.7752551436424255, 0.12487845122814178, 0.7609840631484985, 0.21821437776088715, ...]</code>	<code>Shortterm and current portion of longterm debt</code>	<code>Borrowings</code>
<code>[-0.24947500228881836, 1.0999057292938232, 0.3973265290260315, 0.551521897315979, -0.20870772004127502, ...]</code>	<code>Trade and other</code>	<code>Trade and other payables</code>
<code>[0.16158847510814667, 0.9547826647758484, 0.5619722604751587, 1.3562628030776978, -0.42042723298072815, ...]</code>	<code>Interest rate derivatives</code>	<code>Derivative financial instruments</code>

Loss: <code>MSELoss</code>

en-lv

Dataset: en-lv
Size: 487 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.71 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 8.63 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.5849851369857788, 0.12363594025373459, -0.019146278500556946, 0.223326176404953, 0.3553294241428375, ...]</code>	<code>Noncurrent interestbearing loans</code>	<code>Aiznemumi no kreditiestadem</code>
<code>[-0.4405641555786133, 0.6129574179649353, 0.3001856207847595, 0.2243034392595291, 0.3611409366130829, ...]</code>	<code>Loans long term</code>	<code>Aiznemumi no kreditiestadem</code>
<code>[0.4723680913448334, 0.5573369860649109, -0.02968907356262207, -0.17952217161655426, -0.6545169949531555, ...]</code>	<code>Proceeds from dividends</code>	<code>No meitassabiedribam sanemtas dividendes</code>

Loss: <code>MSELoss</code>

en-el

Dataset: en-el
Size: 104 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 7.91 tokens</li><li>max: 22 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 5.54 tokens</li><li>max: 8 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.4922516345977783, -0.07638876140117645, 0.27244681119918823, -0.03274909406900406, -0.44587045907974243, ...]</code>	<code>other reserves</code>	<code>Reserves</code>
<code>[0.02690565586090088, 0.5322003960609436, -0.22316685318946838, 1.4094343185424805, -1.2200299501419067, ...]</code>	<code>Derivativesliabilities</code>	<code>Derivative financial instruments</code>
<code>[-0.4285869002342224, -1.2929456233978271, -0.05507340282201767, -0.9150614142417908, -1.67551589012146, ...]</code>	<code>Invested unrestricted equity fund</code>	<code>Reserves</code>

Loss: <code>MSELoss</code>

en-et

Dataset: en-et
Size: 136 training samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 10.24 tokens</li><li>max: 25 tokens</li></ul>	<ul><li>min: 7 tokens</li><li>mean: 7.0 tokens</li><li>max: 7 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.23401905596256256, 0.947270393371582, -0.3706150949001312, 0.32394295930862427, -0.10204663872718811, ...]</code>	<code>Depreciation and amortisation including impairment charges</code>	<code>Pohivara kulum</code>
<code>[0.5078503489494324, 0.9610038995742798, 0.028378624469041824, 0.5917476415634155, -1.4292068481445312, ...]</code>	<code>vii</code>	<code>Pohivara kulum</code>
<code>[-0.39173853397369385, 0.42254066467285156, -0.6972977519035339, 0.13764289021492004, 0.11351882666349411, ...]</code>	<code>Total depreciation</code>	<code>Pohivara kulum</code>

Loss: <code>MSELoss</code>

Evaluation Datasets

en-fr

Dataset: en-fr
Size: 27,038 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.77 tokens</li><li>max: 21 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.48 tokens</li><li>max: 21 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Ventes</code>
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Produits des activites ordinaires</code>
<code>[-0.7187896966934204, 0.300822377204895, -0.038356583565473557, 1.0221939086914062, -0.07130642980337143, ...]</code>	<code>Distribution costs</code>	<code>Frais commerciaux</code>

Loss: <code>MSELoss</code>

en-fi

Dataset: en-fi
Size: 7,849 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.34 tokens</li><li>max: 34 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 7.66 tokens</li><li>max: 17 tokens</li></ul>

Samples:

label	english	non_english
<code>[-0.044488366693258286, 0.4498324394226074, 0.35706791281700134, 0.5602209568023682, -0.1801929622888565, ...]</code>	<code>Tax on profit for the year</code>	<code>Tuloverot</code>
<code>[-0.044488366693258286, 0.4498324394226074, 0.35706791281700134, 0.5602209568023682, -0.1801929622888565, ...]</code>	<code>Tax on profit for the year</code>	<code>Income taxes</code>
<code>[-0.10370840132236481, 0.5262670516967773, -0.1583852767944336, 0.05357339233160019, 0.7700905799865723, ...]</code>	<code>Remeasurements of defined benefit plans</code>	<code>Etuuspohjaisen nettovelan uudelleen maarittamisesta johtuvat erat</code>

Loss: <code>MSELoss</code>

en-pl

Dataset: en-pl
Size: 19,308 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.3 tokens</li><li>max: 22 tokens</li></ul>	<ul><li>min: 4 tokens</li><li>mean: 12.43 tokens</li><li>max: 39 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.012203109450638294, 0.6782587766647339, 0.11951778084039688, -0.30175572633743286, -0.6870222091674805, ...]</code>	<code>Administrative expenses</code>	<code>Ogolne koszty administracyjne</code>
<code>[0.11572737991809845, 1.1026246547698975, 0.1337483674287796, 0.13492430746555328, -0.2561548352241516, ...]</code>	<code>Other operating income</code>	<code>Pozostale przychody</code>
<code>[-0.012237715534865856, 0.7524855136871338, 0.0722682923078537, -0.1759086549282074, -0.8265506625175476, ...]</code>	<code>Other operating expenses</code>	<code>Pozostale koszty operacyjne</code>

Loss: <code>MSELoss</code>

en-sv

Dataset: en-sv
Size: 15,902 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.03 tokens</li><li>max: 24 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.03 tokens</li><li>max: 21 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Nettoomsattning</code>
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Summa rorelsens intakter</code>
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Summa intakter</code>

Loss: <code>MSELoss</code>

en-de

Dataset: en-de
Size: 19,441 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.14 tokens</li><li>max: 24 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.0 tokens</li><li>max: 23 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.405586302280426, 0.545492947101593, 0.5445799231529236, 0.5528497695922852, 0.3698521554470062, ...]</code>	<code>Financial income</code>	<code>Finanzertrage</code>
<code>[0.405586302280426, 0.545492947101593, 0.5445799231529236, 0.5528497695922852, 0.3698521554470062, ...]</code>	<code>Financial income</code>	<code>IIIB</code>
<code>[0.10624096542596817, 0.2766471207141876, 0.6653332114219666, 0.09570542722940445, -0.5832860469818115, ...]</code>	<code>Financial expenses</code>	<code>Finanzaufwendungen</code>

Loss: <code>MSELoss</code>

en-it

Dataset: en-it
Size: 15,109 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.12 tokens</li><li>max: 22 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.6 tokens</li><li>max: 20 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.0050657871179282665, 0.7755593061447144, -0.4470928907394409, -0.18634264171123505, 0.390926718711853, ...]</code>	<code>Revenue</code>	<code>Ricavi</code>
<code>[0.11572737991809845, 1.1026246547698975, 0.1337483674287796, 0.13492430746555328, -0.2561548352241516, ...]</code>	<code>Other operating income</code>	<code>Altri proventi</code>
<code>[-0.012237218208611012, 0.7524856925010681, 0.0722685381770134, -0.17590798437595367, -0.8265498876571655, ...]</code>	<code>Other operating expenses</code>	<code>Altri oneri</code>

Loss: <code>MSELoss</code>

en-pt

Dataset: en-pt
Size: 3,206 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.62 tokens</li><li>max: 36 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 6.54 tokens</li><li>max: 12 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.0850653275847435, 0.5872150659561157, 0.3560439944267273, -0.4916071593761444, -0.5272688269615173, ...]</code>	<code>Investments in intangible assets</code>	<code>Ativos intangiveis</code>
<code>[-0.29471272230148315, 0.912581205368042, -0.22577235102653503, 0.051218513399362564, -0.2710682451725006, ...]</code>	<code>Other provisions</code>	<code>Provisoes</code>
<code>[0.03657735511660576, 0.3423381447792053, -0.249881774187088, -0.22646693885326385, 0.7550634145736694, ...]</code>	<code>Remeasurements of defined benefit schemes</code>	<code>Ganhos perdas atuariais</code>

Loss: <code>MSELoss</code>

en-no

Dataset: en-no
Size: 1,541 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.91 tokens</li><li>max: 32 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 5.48 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.11572737991809845, 1.1026246547698975, 0.1337483674287796, 0.13492430746555328, -0.2561548352241516, ...]</code>	<code>Other operating income</code>	<code>Andre driftsinntekter</code>
<code>[0.6171316504478455, 0.09544796496629715, 0.3045019507408142, 1.3532874584197998, -0.5360710024833679, ...]</code>	<code>Net profit for the year</code>	<code>Arets resultat</code>
<code>[0.31753233075141907, 0.9272720813751221, -0.13628403842449188, -0.618966817855835, -0.11626463383436203, ...]</code>	<code>Income tax paid</code>	<code>Betalte skatter</code>

Loss: <code>MSELoss</code>

en-nb

Dataset: en-nb
Size: 1,496 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.35 tokens</li><li>max: 23 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 5.79 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.7072435021400452, 0.33462974429130554, -0.25377699732780457, 0.554284393787384, -0.9292709231376648, ...]</code>	<code>Operating profit EBIT</code>	<code>Resultat etter skatt</code>
<code>[0.6171316504478455, 0.09544817358255386, 0.3045021593570709, 1.3532869815826416, -0.5360713601112366, ...]</code>	<code>Net profit for the year</code>	<code>Resultat etter skatt</code>
<code>[0.6171316504478455, 0.09544817358255386, 0.3045021593570709, 1.3532869815826416, -0.5360713601112366, ...]</code>	<code>Net profit for the year</code>	<code>Resultat</code>

Loss: <code>MSELoss</code>

en-de-de

Dataset: en-de-de
Size: 284 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 8.87 tokens</li><li>max: 27 tokens</li></ul>	<ul><li>min: 5 tokens</li><li>mean: 9.26 tokens</li><li>max: 14 tokens</li></ul>

Samples:

label	english	non_english
<code>[0.005065362900495529, 0.7755594253540039, -0.4470923840999603, -0.1863422989845276, 0.39092710614204407, ...]</code>	<code>Revenue</code>	<code>Umsatzerlose</code>
<code>[0.6505127549171448, 0.502105712890625, 0.05527564138174057, 0.031440261751413345, -0.10601992905139923, ...]</code>	<code>Interest received</code>	<code>Erhaltene Zinsen</code>
<code>[0.5774980783462524, 0.4874580204486847, -0.11888153851032257, 0.025767352432012558, 0.07453231513500214, ...]</code>	<code>Total revenue</code>	<code>Umsatzerlose</code>

Loss: <code>MSELoss</code>

en-es

Dataset: en-es
Size: 12,190 evaluation samples
Columns: <code>label</code>, <code>english</code>, and <code>non_english</code>

Approximate statistics based on the first 1000 samples:

	label	english	non_english
type	list	string	string
details	<ul><li>size: 768 elements</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.24 tokens</li><li>max: 22 tokens</li></ul>	<ul><li>min: 3 tokens</li><li>mean: 9.77 tokens</li><li>max: 23 tokens</li></ul>

Samples:
label english non_english

The model card is truncated. Read the rest on Hugging Face.

Task: Sentence Similarity
Library: sentence-transformers
Type: xlm-roberta
Created: Jun 14, 2024
Updated: Jun 14, 2024