Combined

---
pretty_name: Combined Roleplay Dataset
language:
- en
task_categories:
- text-generation
tags:
- dialogue
- creative-writing
- multiturn
- roleplay
- conversational
configs:
- config_name: chai_prize_reward_model
  data_files:
  - path:
    - chai_prize_reward_model.jsonl.zst
    split: train
- config_name: capybara
  data_files:
  - path:
    - capybara.jsonl.zst
    split: train
- config_name: cakrawala
  data_files:
  - path:
    - cakrawala.jsonl.zst
    split: train
- config_name: creative_writing
  data_files:
  - path:
    - creative_writing.jsonl.zst
    split: train
- config_name: empathetic_dialogues
  data_files:
  - path:
    - empathetic_dialogues.jsonl.zst
    split: train
- config_name: multiturn_mix_1k
  data_files:
  - path:
    - multiturn_mix_1k.jsonl.zst
    split: train
- config_name: multiturn_mix_10k
  data_files:
  - path:
    - multiturn_mix_10k.jsonl.zst
    split: train
- config_name: multiturn_mix_30k
  data_files:
  - path:
    - multiturn_mix_30k.jsonl.zst
    split: train
- config_name: multiturn_mix_v2_1k
  data_files:
  - path:
    - multiturn_mix_v2_1k.jsonl.zst
    split: train
- config_name: multiturn_mix_v2_10k
  data_files:
  - path:
    - multiturn_mix_v2_10k.jsonl.zst
    split: train
- config_name: multiturn_mix_v2_30k
  data_files:
  - path:
    - multiturn_mix_v2_30k.jsonl.zst
    split: train
- config_name: multiturn_mix_v3_1k
  data_files:
  - path:
    - multiturn_mix_v3_k1000.jsonl.zst
    split: train
  default: true
- config_name: multiturn_mix_v3_10k
  data_files:
  - path:
    - multiturn_mix_v3_k10000.jsonl.zst
    split: train
- config_name: multiturn_mix_v3_30k
  data_files:
  - path:
    - multiturn_mix_v3_k30000.jsonl.zst
    split: train
- config_name: soda
  data_files:
  - path:
    - soda.jsonl.zst
    split: train
- config_name: samantha
  data_files:
  - path:
    - guilherme34-samanthadataset-rolesformat.jsonl.zst
    split: train
- config_name: dialogsum
  data_files:
  - path:
    - knkarthick-dialogsum.jsonl.zst
    split: train
- config_name: RPGPT_PublicDomain
  data_files:
  - path:
    - practical-dreamer-RPGPT_PublicDomain.jsonl.zst
    split: train
- config_name: synthetic_characters
  data_files:
  - path:
    - AlekseyKorshuk-synthetic-characters.jsonl.zst
    split: train
- config_name: li2017dailydialog
  data_files:
  - path:
    - li2017dailydialog.jsonl.zst
    split: train
- config_name: Conversational-Reasoning-Topical-Chat
  data_files:
  - path:
    - Conversational-Reasoning-Topical-Chat.jsonl.zst
    split: train
---

# Combined Roleplay Dataset

This dataset combines multi-turn conversations across various AI assistant interactions, creative writing scenarios, and roleplaying exchanges. It aims to improve language models' performance in interactive tasks.

* Multi-turn conversations with a mix of standard AI assistant interactions, creative writing prompts, and roleplays
* English content with a few Spanish, Portuguese, and Chinese conversations
* Conversations limited to 4000 tokens using the Llama 3.1 8B tokenizer (but not done yet for the newest batches)
* Structured format featuring alternating user and AI messages, starting with a system or user prompt and ending with an AI response

## Dataset Structure

1. Multiturn mix:
   - 1K version: 1,000 conversations
   - 10K version: 10,000 conversations
   - 30K version: 30,000 conversations

2. Source datasets:
   - 20231206_chai_prize_reward_model_data: 18,574 lines (only keeping the lines with label = 1)
   - Cakrawala: 13,000 lines
   - Capybara: 15,996 lines
   - CreativeWriting: 8,808 lines
   - Empathetic_dialogues: 19,531 lines
   - SODA: 1,155,128 lines
   - Samantha: 5,868 lines
   - DIALOGSum: 10,883 lines
   - RPGPT_PublicDomain: 3,032 lines
   - Synthetic-characters: 17,668 lines
   - li2017dailydialog: 13,118 lines (all splits merged)
   - Conversational-Reasoning-Topical-Chat: 10,784 lines (all splits merged)

## Dataset Creation

The MultiturnMix data was created by:
1. Randomly sampling 20,000 lines from the SODA dataset and combining them with the other datasets
   - For the v2 and v3 datasets, 13,000 random lines from SODA were selected
2. Embeddings were computed with the [agentlans/snowflake-arctic-embed-xs-zyda-2](https://huggingface.co/agentlans/snowflake-arctic-embed-xs-zyda-2) model
3. Clustering the lines into 1,000, 10,000, or 30,000 k-means clusters to ensure diversity

## Considerations for Using the Data

### Intended Uses
This dataset is primarily intended for training and fine-tuning language models for creative writing, roleplaying, and conversational AI tasks.

### Social Impact and Biases
* The data may exhibit biases in style and formatting due to its synthetic nature.
* It might not represent all genres or fandoms equally well.
* Limited to two-player dialogues, which could differ from more complex multi-party interactions.

### Limitations
* Some subsets may be unsuitable for all audiences.
* English with limited representation of other languages.
* Not comprehensive coverage of creative writing or roleplaying scenarios.
* May still contain poor AI-generated content and repetitive conversations.

## Additional Information

### Dataset Sources
The dataset incorporates data from the following sources:
- [Dampfinchen/Creative_Writing_Multiturn](https://huggingface.co/datasets/Dampfinchen/Creative_Writing_Multiturn)
- [agentlans/Capybara](https://huggingface.co/datasets/agentlans/Capybara)
- [agentlans/allenai-soda](https://huggingface.co/datasets/agentlans/allenai-soda)
- [NarrativAI/CakrawalaRP](https://huggingface.co/datasets/NarrativAI/CakrawalaRP)
- [agentlans/Estwld-empathetic_dialogues_llm](https://huggingface.co/datasets/agentlans/Estwld-empathetic_dialogues_llm)
- [ChaiML/20231206_chai_prize_reward_model_data](https://huggingface.co/datasets/ChaiML/20231206_chai_prize_reward_model_data)
- [Guilherme34/SamanthaDataset-rolesformat](https://huggingface.co/datasets/Guilherme34/SamanthaDataset-rolesformat)
- [agentlans/knkarthick-dialogsum](https://huggingface.co/datasets/agentlans/knkarthick-dialogsum)
- [agentlans/practical-dreamer-RPGPT_PublicDomain](https://huggingface.co/datasets/agentlans/practical-dreamer-RPGPT_PublicDomain)
- [agentlans/AlekseyKorshuk-synthetic-characters](https://huggingface.co/datasets/agentlans/AlekseyKorshuk-synthetic-characters)
- [agentlans/li2017dailydialog](https://huggingface.co/datasets/agentlans/li2017dailydialog)
- [agentlans/Conversational-Reasoning-Topical-Chat](https://huggingface.co/datasets/agentlans/Conversational-Reasoning-Topical-Chat)

### Licensing and Privacy
- The dataset is not known to contain personal or sensitive information
- Users should refer to the original sources for specific licensing information