Affine-RL Dataset
This repository contains datasets for reinforcement learning experiments.
Dataset Summary
| Category | File | Entries |
|---|---|---|
| lgc-v2 | boolean_expressions.jsonl | 100,000 |
| cryptarithm.jsonl | 100,000 | |
| dyck_language.jsonl | 100,000 | |
| dyck_language2.jsonl | 100,000 | |
| dyck_language_reasoning_errors.jsonl | 100,000 | |
| game_of_24.jsonl | 100,000 | |
| operation.jsonl | 100,000 | |
| sudoku.jsonl | 100,000 | |
| lgc-v2 subtotal | 800,000 | |
| openspiel | clobber.jsonl | 8,430 |
| clobber_v2.jsonl | 450 | |
| gin_rummy.jsonl | 4,070 | |
| gin_rummy_v2.jsonl | 200 | |
| goofspiel.jsonl | 9,090 | |
| hex.jsonl | 10,840 | |
| hex_v2.jsonl | 550 | |
| leduc_poker.jsonl | 35,366 | |
| liars_dice.jsonl | 52,000 | |
| liars_dice_v2.jsonl | 52,000 | |
| othello.jsonl | 3,326 | |
| othello_v2.jsonl | 190 | |
| openspiel subtotal | 176,512 | |
| print.jsonl | 100,000 | |
| print subtotal | 100,000 | |
| Total | 1,076,512 |