Dataset Viewer (First 5GB)
Auto-converted to Parquet Duplicate
system
stringclasses
1 value
game_name
stringclasses
1 value
images
images listlengths
2
2
problem
stringclasses
1 value
answers
stringclasses
67 values
action
stringclasses
4 values
reward
sequencelengths
1
1
prev_state
unknown
current_state
unknown
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
LEFT
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 151, 213,...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
LEFT
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 87, 164, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
FIRE
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 87, 164, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 87, 164, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
RIGHT
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 87, 164, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
FIRE
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
FIRE
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
You are an advanced Vision-Language Model (VLM) designed to play video games which will be given with 2 game images if available, and you need to analysis the previous game image and current game image to determine the next move. Each game is skipped by 4 frames. Follow the steps below to determine the optimal move: 1...
BreakoutNoFrameskip-v4
You are playing a Video game. It is your turn to move. What is the next move?
<think> First, I need to reason about the game state by examining each category one by one. </think> <think> Agent Location Category: paddle is at bottom left. </think> <think> Small Object Location Category: ball is at top left. </think> <think> Other Object Location Category: no relevant objects present. </think> <t...
FIRE
[ 0 ]
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
[ 128, 4, 149, 88, 30, 0, 0, 0, 0, 0, 0, 140, 14, 97, 108, 101, 95, 112, 121, 46, 95, 97, 108, 101, 95, 112, 121, 148, 140, 8, 65, 76, 69, 83, 116, 97, 116, 101, 148, 147, 148, 41, 129, 148, 66, 45, 30, 0, 0, 247, 60, ...
"You are an advanced Vision-Language Model (VLM) designed to play video games which will be given wi(...TRUNCATED)
BreakoutNoFrameskip-v4
[{"src":"https://datasets-server.huggingface.co/assets/liyang619/atari_breakout_rl_state_500k/--/{da(...TRUNCATED)
You are playing a Video game. It is your turn to move. What is the next move?
"<think> First, I need to reason about the game state by examining each category one by one. </think(...TRUNCATED)
LEFT
[ 0 ]
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAPc8BgD3PAYAPgAAAD4AAAAAAAAAAAAAAAk(...TRUNCATED)
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAFekBwD3PAYAQgAAAEIAAAAAAAAAAAAAAAk(...TRUNCATED)
"You are an advanced Vision-Language Model (VLM) designed to play video games which will be given wi(...TRUNCATED)
BreakoutNoFrameskip-v4
[{"src":"https://datasets-server.huggingface.co/assets/liyang619/atari_breakout_rl_state_500k/--/{da(...TRUNCATED)
You are playing a Video game. It is your turn to move. What is the next move?
"<think> First, I need to reason about the game state by examining each category one by one. </think(...TRUNCATED)
NOOP
[ 0 ]
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAFekBwD3PAYAQgAAAEIAAAAAAAAAAAAAAAk(...TRUNCATED)
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAFekBwD3PAYARgAAAEYAAAAAAAAAAAAAAAk(...TRUNCATED)
"You are an advanced Vision-Language Model (VLM) designed to play video games which will be given wi(...TRUNCATED)
BreakoutNoFrameskip-v4
[{"src":"https://datasets-server.huggingface.co/assets/liyang619/atari_breakout_rl_state_500k/--/{da(...TRUNCATED)
You are playing a Video game. It is your turn to move. What is the next move?
"<think> First, I need to reason about the game state by examining each category one by one. </think(...TRUNCATED)
RIGHT
[ 0 ]
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAFekBwD3PAYARgAAAEYAAAAAAAAAAAAAAAk(...TRUNCATED)
"gASVWB4AAAAAAACMDmFsZV9weS5fYWxlX3B5lIwIQUxFU3RhdGWUk5QpgZRCLR4AAPc8BgD3PAYASgAAAEoAAAAAAAAAAAAAAAk(...TRUNCATED)
End of preview. Expand in Data Studio

No dataset card yet

Downloads last month
4