metadata

dataset_info:
  features:
    - name: conversations
      list:
        - name: from
          dtype: string
        - name: value
          dtype: string
  splits:
    - name: train
      num_bytes: 916499
      num_examples: 1149
  download_size: 91505
  dataset_size: 916499
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Rick and Morty Transcripts - ShareGPT-3

Overview

The Mohaaxa/rick-and-morty-transcripts-sharegpt-3 dataset is designed for fine-tuning language models to emulate the character Morty from Rick and Morty. It is derived from the Prarabdha/Rick_and_Morty_Transcript, which itself was inspired by the Rick & Morty Scripts by Andrada Olteanu.

This dataset expands on previous versions by including updated content and improved structure, making it a valuable resource for fine-tuning conversational models.

Inspiration

The dataset was inspired by the original Rick & Morty scripts by Andrada Olteanu but aims to provide a larger and more up-to-date collection of transcripts. It covers nearly all episodes up to Season 5.

Dataset Details

Rows: 1.15k
Columns:
- index: Row index
- speaker: Character's name (e.g., Rick, Morty)
- dialogue: The spoken dialogue for that entry

Usage

This dataset is ideal for:

Fine-tuning LLMs to emulate Morty
Generating character-specific dialogues
Exploring natural language understanding in scripted conversations

Example Usage (Hugging Face Datasets Library)

from datasets import load_dataset

dataset = load_dataset("Mohaaxa/rick-and-morty-transcripts-sharegpt-3")
print(dataset["train"][0])