Oasst2 Es 3K Topics: Spanish Short Conversations with Topic Labels

A dataset of short Spanish conversations derived from thinkPy/oasst2-short-es. It includes two columns of automatically assigned topic classifications generated using the MoritzLaurer/mDeBERTa-v3-base-mnli-xnli model. The dataset was created by thinkPy and was last updated on 2026-05-31.

Use Cases

• Fine-tuning topic classification models based on the provided topic labels.
• Benchmarking Spanish language understanding models on conversation data.
• Analyzing conversation themes and structures in Spanish dialogue datasets.
• Training or evaluating multilingual NLP pipelines that include Spanish.

Strengths

• Topic labels are generated using a specific, named NLP model (MoritzLaurer/mDeBERTa-v3-base-mnli-xnli).
• Derived from a known base dataset (OpenAssistant/oasst2) via a Spanish-language version (thinkPy/oasst2-short-es).

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

Derived from thinkPy/oasst2-short-es, which is based on OpenAssistant/oasst2.

Collection Method

A random sample (seed=42) was taken from the base dataset and augmented with two topic classification columns.

Freshness

Last updated 2026-05-31 13:20:18; freshness should be verified.