metadata
annotations_creators: []
language:
- hi
- en
- multilingual
language_creators: []
license: []
multilinguality:
- multilingual
pretty_name: adaption-multilingual_instruction_tuning
size_categories:
- 10K<n<100K
source_datasets:
- extended|https://huggingface.co/datasets/deepmage121/sharegpt-mix-20k
tags:
- adaption
- instruction-tuning
- code
- writing-editing-communication
- science
task_categories: []
task_ids: []
This dataset is a remastered version of this dataset prepared using Adaption's Adaptive Data platform.
adaption-multilingual_instruction_tuning
This dataset contains multi-turn conversational pairs featuring instructions and responses in Bengali, Hindi, and English. The content spans diverse domains including creative storytelling, code generation, technical troubleshooting, legal documentation, and conceptual explanations tailored for different audiences. Each sample demonstrates an assistant following user prompts to translate, expand, debug, or explain complex topics.
Dataset size
There are 40,880 data points in this dataset. This is an instruction tuning dataset.
Quality of Remastered Dataset
The final quality is B, with a relative quality improvement of 40.0%.
Domain
- Code (34%)
- Writing-editing-communication (12%)
- Science (10%)
Language
- Hindi (30%)
- Bangla (26%)
- English (24%)
Tone
- Technical (22%)
- Professional (16%)
- Explanatory (12%)
Evaluation Results
Quality Gains:
Grade Improvement:
Percentile Chart:
