Oasst1-21k-Ja: Japanese Instruction Tuning Dataset for LLMs

LLM-jp, a collaborative project in Japan, provides this dataset. It is a Japanese translation of a 21,000-instruction English subset from the OASST1 dataset, created using the DeepL translation service. The dataset was last updated on February我们发现一个错误。根据输入，数据集标题是“Oasst1 21K Ja”，描述中提到它是“oasst1-21k-ja”，并说明是“Japanese translation of an English subset of oasst1”。因此，正确的摘要应基于此信息。输入中没有明确的行数“21,000”，但标题和名称暗示了“21k”。我将据此修正摘要。

Use Cases

• Fine-tuning Japanese language models for instruction-following based on the translated instruction-response pairs.
• Evaluating the performance of LLMs on Japanese conversational tasks.
• Studying the effects of machine translation on instruction-tuning data quality.

Strengths

• Dataset is specifically designed for Japanese language model instruction tuning.
• Translation was performed using the DeepL service, which may imply a degree of quality control.
• The source is a known instruction dataset (OASST1), providing a foundation for the content.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• Data may reflect translation bias or artifacts introduced by the machine translation process.

Provenance

Source

LLM-jp collaborative project.

Collection Method

Machine translation of an English subset from the OASST1 dataset using DeepL.

Time Range

null

Freshness

Last updated 2024-02-06 04:06:04; freshness should be verified.

Geography

Japan (by project origin and language focus).