README.md

---
dataset_info:
- config_name: agent_profiles
  features:
  - name: pk
    dtype: string
  - name: first_name
    dtype: string
  - name: last_name
    dtype: string
  - name: age
    dtype: int64
  - name: occupation
    dtype: string
  - name: gender
    dtype: string
  - name: gender_pronoun
    dtype: string
  - name: public_info
    dtype: string
  - name: big_five
    dtype: string
  - name: moral_values
    list: string
  - name: schwartz_personal_values
    list: string
  - name: personality_and_values
    dtype: string
  - name: decision_making_style
    dtype: string
  - name: secret
    dtype: string
  - name: model_id
    dtype: string
  - name: mbti
    dtype: string
  splits:
  - name: train
    num_bytes: 23439
    num_examples: 40
  download_size: 21939
  dataset_size: 23439
- config_name: env_agent_combos
  features:
  - name: pk
    dtype: string
  - name: env_id
    dtype: string
  - name: agent_ids
    list: string
  splits:
  - name: train
    num_bytes: 55800
    num_examples: 450
  download_size: 42172
  dataset_size: 55800
- config_name: environment_profiles
  features:
  - name: pk
    dtype: string
  - name: codename
    dtype: string
  - name: source
    dtype: string
  - name: scenario
    dtype: string
  - name: agent_goals
    list: string
  - name: relationship
    dtype: int64
  - name: age_constraint
    dtype: string
  - name: occupation_constraint
    dtype: string
  - name: agent_constraint
    dtype: string
  - name: task_type
    dtype: string
  splits:
  - name: train
    num_bytes: 82619
    num_examples: 90
  download_size: 74278
  dataset_size: 82619
- config_name: environment_profiles_synthetic
  features:
  - name: pk
    dtype: 'null'
  - name: codename
    dtype: string
  - name: source
    dtype: string
  - name: scenario
    dtype: string
  - name: agent_goals
    list: string
  - name: relationship
    dtype: int64
  - name: age_constraint
    dtype: string
  - name: occupation_constraint
    dtype: string
  - name: agent_constraint
    dtype: 'null'
  - name: task_type
    dtype: string
  splits:
  - name: train
    num_bytes: 250596
    num_examples: 360
  download_size: 179737
  dataset_size: 250596
- config_name: episodes
  features:
  - name: pk
    dtype: string
  - name: environment
    dtype: string
  - name: agents
    dtype: string
  - name: tag
    dtype: string
  - name: models
    dtype: string
  - name: messages
    dtype: string
  - name: reasoning
    dtype: string
  - name: rewards
    dtype: string
  - name: rewards_prompt
    dtype: string
  splits:
  - name: train
    num_bytes: 461313109
    num_examples: 21333
  download_size: 109572239
  dataset_size: 461313109
- config_name: relationship_profiles
  features:
  - name: pk
    dtype: string
  - name: agent_1_id
    dtype: string
  - name: agent_2_id
    dtype: string
  - name: relationship
    dtype: int64
  - name: background_story
    dtype: string
  splits:
  - name: train
    num_bytes: 55725
    num_examples: 120
  download_size: 28267
  dataset_size: 55725
configs:
- config_name: agent_profiles
  data_files:
  - split: train
    path: agent_profiles/train-*
- config_name: env_agent_combos
  data_files:
  - split: train
    path: env_agent_combos/train-*
- config_name: environment_profiles
  data_files:
  - split: train
    path: environment_profiles/train-*
- config_name: environment_profiles_synthetic
  data_files:
  - split: train
    path: environment_profiles_synthetic/train-*
- config_name: episodes
  data_files:
  - split: train
    path: episodes/train-*
- config_name: relationship_profiles
  data_files:
  - split: train
    path: relationship_profiles/train-*
---