SnakModel: Lessons Learned from Training an Open Danish Large Language Model

• URL: http://arxiv.org/abs/2412.12956v1
• Date: Tue, 17 Dec 2024 14:38:21 GMT
• Title: SnakModel: Lessons Learned from Training an Open Danish Large Language Model
• Authors: Mike Zhang, Max Müller-Eberstein, Elisa Bassignana, Rob van der Goot,
• Abstract summary: We present SnakModel, a Danish large language model (LLM) based on Llama2-7B.<n>We continuously pre-train on 13.6B Danish words, and further tune on 3.7M Danish instructions.
• Score: 18.6197271443555
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We present SnakModel, a Danish large language model (LLM) based on Llama2-7B, which we continuously pre-train on 13.6B Danish words, and further tune on 3.7M Danish instructions. As best practices for creating LLMs for smaller language communities have yet to be established, we examine the effects of early modeling and training decisions on downstream performance throughout the entire training pipeline, including (1) the creation of a strictly curated corpus of Danish text from diverse sources; (2) the language modeling and instruction-tuning training process itself, including the analysis of intermediate training dynamics, and ablations across different hyperparameters; (3) an evaluation on eight language and culturally-specific tasks. Across these experiments SnakModel achieves the highest overall performance, outperforming multiple contemporary Llama2-7B-based models. By making SnakModel, the majority of our pre-training corpus, and the associated code available under open licenses, we hope to foster further research and development in Danish Natural Language Processing, and establish training guidelines for languages with similar resource constraints.

Related papers

• SLAM: Towards Efficient Multilingual Reasoning via Selective Language Alignment [78.4550589538805]
  We propose an efficient multilingual reasoning alignment approach that precisely identifies and fine-tunes the layers responsible for handling multilingualism. Experimental results show that our method, SLAM, only tunes 6 layers' feed-forward sub-layers including 6.5-8% of all parameters within 7B and 13B LLMs.
  arXiv  Detail & Related papers  (2025-01-07T10:29:43Z)
• A Comparison of Language Modeling and Translation as Multilingual Pretraining Objectives [13.581385765600265]
  Pretrained language models (PLMs) display impressive performances and have captured the attention of the NLP community. This paper proposes a comparison of multilingual pretraining objectives in a controlled methodological environment.
  arXiv  Detail & Related papers  (2024-07-22T09:16:30Z)
• CroissantLLM: A Truly Bilingual French-English Language Model [42.03897426049679]
  We introduce CroissantLLM, a 1.3B language model pretrained on a set of 3T English and French tokens. We pioneer the approach of training an intrinsically bilingual model with a 1:1 English-to-French pretraining data ratio. To assess performance outside of English, we craft a novel benchmark, FrenchBench.
  arXiv  Detail & Related papers  (2024-02-01T17:17:55Z)
• Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research [139.69207791947738]
  Dolma is a three-trillion-token English corpus built from a diverse mixture of web content, scientific papers, code, public-domain books, social media, and encyclopedic materials. We document Dolma, including its design principles, details about its construction, and a summary of its contents. We present analyses and experimental results on intermediate states of Dolma to share what we have learned about important data curation practices.
  arXiv  Detail & Related papers  (2024-01-31T20:29:50Z)
• PolyLM: An Open Source Polyglot Large Language Model [57.64420154135178]
  We present PolyLM, a multilingual large language model (LLMs) trained on 640 billion (B) tokens, avaliable in two model sizes: 1.7B and 13B. To enhance its multilingual capabilities, we 1) integrate bilingual data into training data; and 2) adopt a curriculum learning strategy that increases the proportion of non-English data from 30% in the first stage to 60% in the final stage during pre-training. Further, we propose a multilingual self-instruct method which automatically generates 132.7K diverse multilingual instructions for model fine-tuning.
  arXiv  Detail & Related papers  (2023-07-12T09:00:37Z)
• Pre-Training to Learn in Context [138.0745138788142]
  The ability of in-context learning is not fully exploited because language models are not explicitly trained to learn in context. We propose PICL (Pre-training for In-Context Learning), a framework to enhance the language models' in-context learning ability. Our experiments show that PICL is more effective and task-generalizable than a range of baselines, outperforming larger language models with nearly 4x parameters.
  arXiv  Detail & Related papers  (2023-05-16T03:38:06Z)
• CodeGen2: Lessons for Training LLMs on Programming and Natural Languages [116.74407069443895]
  We unify encoder and decoder-based models into a single prefix-LM. For learning methods, we explore the claim of a "free lunch" hypothesis. For data distributions, the effect of a mixture distribution and multi-epoch training of programming and natural languages on model performance is explored.
  arXiv  Detail & Related papers  (2023-05-03T17:55:25Z)
• Generalizing Multimodal Pre-training into Multilingual via Language Acquisition [54.69707237195554]
  English-based Vision-Language Pre-training has achieved great success in various downstream tasks. Some efforts have been taken to generalize this success to non-English languages through Multilingual Vision-Language Pre-training. We propose a textbfMultitextbfLingual textbfAcquisition (MLA) framework that can easily generalize a monolingual Vision-Language Pre-training model into multilingual.
  arXiv  Detail & Related papers  (2022-05-29T08:53:22Z)
• Pre-training Data Quality and Quantity for a Low-Resource Language: New Corpus and BERT Models for Maltese [4.4681678689625715]
  We analyse the effect of pre-training with monolingual data for a low-resource language. We present a newly created corpus for Maltese, and determine the effect that the pre-training data size and domain have on the downstream performance. We compare two models on the new corpus: a monolingual BERT model trained from scratch (BERTu), and a further pre-trained multilingual BERT (mBERTu)
  arXiv  Detail & Related papers  (2022-05-21T06:44:59Z)
• Training dataset and dictionary sizes matter in BERT models: the case of Baltic languages [0.0]
  We train a trilingual LitLat BERT-like model for Lithuanian, Latvian, and English, and a monolingual Est-RoBERTa model for Estonian. We evaluate their performance on four downstream tasks: named entity recognition, dependency parsing, part-of-speech tagging, and word analogy.
  arXiv  Detail & Related papers  (2021-12-20T14:26:40Z)
• Cross-lingual Transferring of Pre-trained Contextualized Language Models [73.97131976850424]
  We propose a novel cross-lingual model transferring framework for PrLMs: TreLM. To handle the symbol order and sequence length differences between languages, we propose an intermediate TRILayer" structure. We show the proposed framework significantly outperforms language models trained from scratch with limited data in both performance and efficiency.
  arXiv  Detail & Related papers  (2021-07-27T06:51:13Z)
• HerBERT: Efficiently Pretrained Transformer-based Language Model for Polish [4.473327661758546]
  This paper presents the first ablation study focused on Polish, which, unlike the isolating English language, is a fusional language. We design and thoroughly evaluate a pretraining procedure of transferring knowledge from multilingual to monolingual BERT-based models. Based on the proposed procedure, a Polish BERT-based language model -- HerBERT -- is trained.
  arXiv  Detail & Related papers  (2021-05-04T20:16:17Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.