This model is a continually pretrained version of the google/gemma-2-2b architecture, fine-tuned on extensive Bangla datasets. The primary goal of the continual pretraining was to enhance the model's ability to generate high-quality Bangla text. By extending the pretraining process specifically on Bangla data, the model has demonstrated superior performance in Bangla language understanding evaluation benchmarks and text generation tasks.
Model Architecture: Gemma 2 is an auto-regressive language model with optimized transformer architecture.
Training Data
Params
Input modalities
Output modalities
Context Length
Token count
Gemma 2
Hishab curated Bangla text corpus
2B
Monolingual Text(Bangla)
Monolingual Text(Bangla)
4096
3B tokens
How To Use
Below we share some code snippets on how to get quickly started with running the model. First, install the Transformers library with:
pip install -U transformers
Then, copy the snippet from the section that is relevant to your use case.
Training Factors: We used llama-factory training library, cloud GPU cluster, and production infrastructure for pretraining. Fine-tuning, annotation, and evaluation were also performed on cloud infrastructure.
Training Data
Overview: We have collected a large Bangla raw dataset of text data from a wide variety of sources. Our collected data so far includes a mix of web documents, books, translated text, transliterated text, transcribed text, code-mixed text, conversations, and open-source raw data. The dataset is cleaned and filtered by different filtering criteria to ensure the quality of the data. Our collected data size is roughly around 268 GB. We separated 22GB data from that using a ratio of the actual data size. Total trained tokens are 3B tokens.
Data sources summary:
Web documents: Extracted, clean, and filtered common crawl data
Books: Extracted, clean, filtered books data
Transcribed text: Used in-house Bangla ASR model to transcribe Bangla audio data
Translation data: We trained an English-Bangla translation LLM model and used it to translate English data to Bangla
Code-mixed data: We trained an English-Bangla code-mixed LLM model and used it to generate code-mixed data
Transliteration data: We trained a Bangla-English transliteration LLM model and used it to generate transliterated data
Synthetic data: We generated synthetic data using a Bangla LLM model
Others: We scrapped some selected website data, used open-source data, and used some other data sources
Benchmarks
In this section, we report the results for titulm-gemma-2-2b-v1.0 models on standard automatic benchmarks. For all these evaluations, we used lm-evaluation-harness evaluations library.
Evaluation Datasets
We evaluated our pre-trained models on both Bangla and English benchmark datasets. Although the model is trained on Bangla data, its English capability is also evaluated on English benchmark datasets. The evaluation datasets are as follows:
Bangla Benchmark datasets
We evaluated the models on the following datasets:
Bangla MMLU: A private multiple choice question dataset developed by Hishab curated from various sources.
CommonsenseQa Bangla: A Bangla translation of the CommonsenseQA dataset. The dataset was translated using a new method called Expressive Semantic Translation (EST), which combines Google Machine Translation with LLM-based rewriting modifications.
OpenbookQA Bangla: A Bangla translation of the OpenbookQA dataset. The dataset was translated using a new method called Expressive Semantic Translation (EST), which combines Google Machine Translation with LLM-based rewriting modifications.
Piqa Bangla: A Bangla translation of the Piqa dataset. The dataset was translated using a new method called Expressive Semantic Translation (EST), which combines Google Machine Translation with LLM-based rewriting modifications.
BoolQ Bangla: The dataset contains 15,942 examples, with each entry consisting of a triplet: (question, passage, answer). The questions are naturally occurring, generated from unprompted and unconstrained settings. Input passages were sourced from Bangla Wikipedia, Banglapedia, and News Articles, and GPT-4 was used to generate corresponding yes/no questions with answers.
English Benchmark datasets
MMLU: This is a massive multitask test consisting of multiple-choice questions from various branches of knowledge.
CommonseQa: CommonsenseQA is a new multiple-choice question-answering dataset that requires different types of commonsense knowledge to predict the correct answers .
OpenbookQA: OpenBookQA aims to promote research in advanced question-answering, probing a deeper understanding of both the topic (with salient facts summarized as an open book, also provided with the dataset) and the language it is expressed in.
Piqa: The PIQA dataset focuses on physical commonsense reasoning, challenging AI to handle everyday situations requiring practical knowledge and unconventional solutions. Inspired by instructables.com, it aims to enhance AI's ability to understand and reason about physical interactions.
BoolQ: BoolQ is a question-answer dataset for yes/no questions containing 15942 examples. These questions are naturally occurring. They are generated in unprompted and unconstrained settings. Each example is a triplet of (question, passage, answer), with the title of the page as optional additional context. The text-pair classification setup is similar to existing natural language inference tasks.
Evaluation Results
Evaluation on Bangla Benchmark datasets
gemma-2-2b performs better in Bangla MMLU and BoolQ BN in the 0-shot setting.
titulm-gemma-2-2b-v1.0 outperforms in Commonsense QA BN, OpenBook QA BN, and PIQA BN across both 0-shot and 5-shot settings.
In the 5-shot setting, titulm-gemma-2-2b-v1.0 achieves the highest scores in BoolQ BN, Commonsense QA BN, and OpenBook QA BN.
PIQA BN shows consistent performance across both models, with titulm-gemma-2-2b-v1.0 leading in both settings.
Model
Shots
Bangla MMLU
BoolQ BN
Commonsense QA BN
OpenBook QA BN
PIQA BN
gemma-2-2b
0-shot
0.32
0.63
0.26
0.34
0.56
5-shot
0.35
0.46
0.28
0.33
0.56
titulm-gemma-2-2b-v1.0
0-shot
0.31
0.59
0.31
0.36
0.63
5-shot
0.35
0.59
0.41
0.37
0.62
Evaluation on English Benchmark datasets
gemma-2-2b outperforms titulm-gemma-2-2b-v1.0 across all tasks in both 0-shot and 5-shot settings, achieving the highest scores in MMLU, BoolQ, Commonsense QA, OpenBook QA, and PIQA, with a peak 5-shot score of 0.80 in PIQA.
titulm-gemma-2-2b-v1.0 shows competitive performance but lags behind gemma-2-2b, particularly in Commonsense QA and BoolQ, with the highest score being 0.77 in PIQA.
It is expected as we have trained our model only on Bangla text.
