Dataset Viewer
Auto-converted to Parquet Duplicate
episode_id
int64
0
511
env_id
stringclasses
2 values
model_name
stringclasses
1 value
opponent_name
stringclasses
1 value
completion_status
stringclasses
2 values
model_outcome
stringclasses
2 values
model_reward
float64
-1
1
opponent_reward
float64
-1
1
num_turns
int64
2
25
duration
float64
0.95
156
trajectory
listlengths
46
95
observations
listlengths
46
95
formatted_observations
listlengths
46
95
reasoning
listlengths
46
95
actions
listlengths
46
95
player_ids
listlengths
46
95
model_names
listlengths
46
95
26
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
invalid
loss
-1
0
8
55.823801
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":26,"final_reward":0.0,"formatted_observation":(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nOb(...TRUNCATED)
["","","Observation: \n[GAME] You are Player 1 in a 5 round game of Kuhn Poker.\nGame Rules:\n- Kuhn(...TRUNCATED)
["[4]","Hello! I'm interested in trading to improve the value of my resources. I have 10 Wood (value(...TRUNCATED)
[0,0,1,0,0,0,0,0,1,1,1,1,1,1,1,1,1,0,0,0,0,0,1,0,1,0,0,1,0,1,1,0,1,0,1,1,0,1,1,1,0,0,1,0,0,1,1,0,0,1(...TRUNCATED)
["openai/gpt-4.1-mini","openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","openai/gpt-4(...TRUNCATED)
7
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
invalid
win
0
-1
3
9.450787
[{"action":"","env_id":"TicTacToe-v0","episode_id":7,"final_reward":-1.0,"formatted_observation":"<|(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObservat(...TRUNCATED)
["As Player 0, your goal is to win the game of Tic Tac Toe by placing three of your marks ('O') in a(...TRUNCATED)
["","[check]","Hello! I see you have Wood and Gold. I currently have the following resources and val(...TRUNCATED)
[0,1,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,0,1,1,1,1,0,0,0,0,0,0,1,1,0,1,0,1,1,1,1,0,1,0,0,1,0,0,1,0,0,1,0,1(...TRUNCATED)
["qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","openai/gpt-4.1-mini","openai/gpt-4.1-mini","openai/gpt-(...TRUNCATED)
495
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
win
1
-1
7
23.167705
[{"action":"4","env_id":"TicTacToe-v0","episode_id":495,"final_reward":1.0,"formatted_observation":"(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObservat(...TRUNCATED)
["To determine the best move, I'll consider the current board state and available moves. Here's my r(...TRUNCATED)
["4","[bet]","[right]/[d]","","[col 3]","[5]","[Bid: 2, 4]","[0 2]","[0]","[fold]","[up]","Hello Pla(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,0,0,0,1,1,1,1,1,1,1,1,0,0,0,0,0,0,1,0,0,1,1,1,1,1,1,1,0,0(...TRUNCATED)
["qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","openai/gpt-4.(...TRUNCATED)
14
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
invalid
win
0
-1
3
18.835394
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":14,"final_reward":1.0,"formatted_observation":(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","Your card is J, which is the lowest card in the deck. In Kuhn Poker, the player with the lowest(...TRUNCATED)
["[4]","[bet]","[right]/[d]","","[col 3]","[5]","[Bid: 3, 4]","[0 2]","1","[call]","[d]","Hello! I s(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,0,1,1,1,1,0,1,0,0,1,0,1,0,1,1,1,0,1,0,0,1,0,0,1,0,1,1,0,1(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","qwen/Qwen3-8B(...TRUNCATED)
453
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
win
1
-1
6
14.809003
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":453,"final_reward":0.0,"formatted_observation"(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nOb(...TRUNCATED)
["","","To win this game, I need to maximize the value of my resources while ensuring that my oppone(...TRUNCATED)
["[4]","[check]","[Offer: 2 Wood -> 1 Gold]","[down]/[s]","[col 3]","[0]","[Bid: 1, 1]","[2 5]","","(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,1,1,0,0,0,0,0,0,0,0,1,1,1,1,1,0,1,0,0,0,1,0,1,0,1,1,0,1,1,0,1,0,1,0(...TRUNCATED)
["openai/gpt-4.1-mini","openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","qwen/Qwen3-8(...TRUNCATED)
510
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
loss
-1
1
5
27.631853
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":510,"final_reward":-1.0,"formatted_observation(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","Let's play this game of Kuhn Poker step by step, reasoning through each decision carefully.\n\n(...TRUNCATED)
["[4]","[bet]","\n\"Hello! It's nice to meet you in this game. I'm looking forward to negotiating wi(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,0,1,1,1,1,0,0,0,1,0,0,1,0,1,1,1,0,0,1,0,1,0,1,0,1,1,0,1,0,1,1,1,0,0,1,1,0,0,1(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","qwen/Qwen3-8B(...TRUNCATED)
29
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
loss
-1
1
4
7.161719
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":29,"final_reward":1.0,"formatted_observation":(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","### Step-by-Step Reasoning:\n\n1. **Understanding the Game State:**\n - **Round 1 of 5:** Thi(...TRUNCATED)
["[4]","[bet]","Great! Here's your 3 Gold.","[d]","[col 3]","13","[Bid: 3, 2]","[2 5]","0","[call]",(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,1,1,1,1,1,1,1,0,0,0,1,0,0,0,1,0,1,1,1,1,0,0,0,0,1,1,1,0,0(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","qwen/Qwen3-8(...TRUNCATED)
284
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
invalid
win
0
-1
4
37.372499
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":284,"final_reward":0.0,"formatted_observation"(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","Let's analyze the situation step by step:\n\n1. **Game State**: We are in round 1 of a 5-round (...TRUNCATED)
["[4]","[bet]","[Offer: 3 Wood -> 2 Gold] \"I'm looking to increase the value of my resources, so I'(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,0,1,1,0,1,0,0,1,0,1,1,1,1,0,1,0,1,1,0,1,0,1,0,1,0,1,1,0,0,0,0,1,1,1,1,1,1,0(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","qwen/Qwen3-8(...TRUNCATED)
5
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
win
1
-1
7
60.053326
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":5,"final_reward":-1.0,"formatted_observation":(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","The initial hand strength is moderate, with a Queen. The game has a low betting limit and a hig(...TRUNCATED)
["[4]","[check]","[d]","[col 3]","","[5]","[Bid: 3, 2]","[2 5]","0","[check]","[s]","[col 3]","Hi, I(...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,1,0,0,1,1,1,0,1,0,1,1,0,1,1,0,1,0,1,1,0,0,1,0,0,0,0,1,0,1(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8(...TRUNCATED)
192
TicTacToe-v0
qwen/Qwen3-8B-Base
openai/gpt-4.1-mini
complete
loss
-1
1
3
6.762016
[{"action":"[4]","env_id":"TicTacToe-v0","episode_id":192,"final_reward":1.0,"formatted_observation"(...TRUNCATED)
["You are Player 0 in Tic Tac Toe.\nYour goal is to win three in a row (horizontally, vertically, or(...TRUNCATED)
["","<|im_start|>user\nYou are playing a two-player zero-sum game. Make valid actions to win.\nObser(...TRUNCATED)
["","Let's break down the situation step by step:\n\n1. **Current Round**: Round 1 out of 5 rounds.\(...TRUNCATED)
["[4]","[bet]","","[d]","[col 3]","[5]","[Bid: 3, 4]","[2 5]","[1]","[call]","Hello Player 0! I see (...TRUNCATED)
[0,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,0,1,0,1,1,1,1,1,1,0,0,0,0,0,0,0,1,1,1,1,1,1,1,0,0,0,0,0,0(...TRUNCATED)
["openai/gpt-4.1-mini","qwen/Qwen3-8B-Base","qwen/Qwen3-8B-Base","openai/gpt-4.1-mini","qwen/Qwen3-8(...TRUNCATED)
End of preview. Expand in Data Studio
README.md exists but content is empty.
Downloads last month
4