Datasets:

---

OpenAssistant

/

oasst1

like 1.45k

Follow

OpenAssistant 1.05k

Reveals the interaction between prompts and completions in Indonesian language, providing insight into the dialogue flow and content.

Provides a random sample of prompt-completion pairs in English, which can be useful for evaluating the quality and coherence of assistant responses in a conversational dataset.

Retrieves pairs of prompts and their corresponding completions in English, providing insight into the interaction patterns between users and assistants.

Retrieves unique prompt-completion pairs from the dataset, providing a clear view of the interactions between prompts and their corresponding assistant responses.

Provides a meaningful sample of user queries and corresponding assistant responses in English, helping to understand the nature of interactions and the quality of responses in the dataset.

This query pairs user queries with corresponding assistant responses, filtering out very short messages, which can help understand the dynamics of user-assistant interactions.

Provides a paired list of user queries and corresponding assistant responses, useful for understanding the interaction patterns in a conversational dataset.

Reveals example dialogues by pairing user queries with corresponding assistant responses, providing insight into how the assistant interacts with users.

Reveals multi-turn conversation patterns, showing how user queries and assistant responses evolve over multiple exchanges, providing useful insights into dialogue dynamics.

Reveals a sequence of interactions between an assistant and a user, along with the assistant's final response, helping to understand the flow and context of the conversation.

Retrieves English text entries shorter than 100 characters that are parent posts, providing basic filtering of the dataset.

Retrieves records from the train dataset where the parent_id is NULL, providing a basic filter but limited analytical value.

Retrieves English prompts from the dataset that do not have a parent, providing a basic view of initial prompts in English.

Lists all unique languages present in the dataset, providing basic information about language diversity.

Retrieves all English entries without a parent ID, providing a basic filtered view of the dataset.

Retrieves all English entries without a parent, providing a basic filtered view of the dataset.

Retrieves all records from the 'train' dataset where the language is Arabic, providing a basic filter of the data.

Retrieves all entries where the role is 'prompter' and the language is French, providing a basic filtered view of the dataset.

This query retrieves all entries for a specific user in the Russian language, providing basic information but limited analytical value.

This query returns a small sample of Spanish-language assistant responses under 200 characters, which provides basic filtering but limited insight into the dataset's patterns.

Retrieves non-synthetic text entries that do not have a parent, useful for identifying standalone entries.

The query retrieves English prompts without a parent ID, which provides basic filtering but minimal analytical insight.

Returns a set of prompt-answer pairs, which provides a basic view of interactions but does not offer deeper analytical insights.

The query retrieves 5000 prompt-answer pairs in English where the prompt is directly followed by an assistant's response. This provides a basic glimpse into the conversation dynamics but does not reveal significant trends or insights.