Giters
MARIO-Math-Reasoning
/
Super_MARIO
Geek Repo:
Geek Repo
Github PK Tool:
Github PK Tool
Stargazers:
196
Watchers:
13
Issues:
18
Forks:
14
MARIO-Math-Reasoning/Super_MARIO Issues
Problem running evaluation
Updated
12 days ago
Comments count
1
Issue about requirements file
Closed
a month ago
Comments count
3
About Training data generation.
Closed
a month ago
Comments count
2
type of template for training
Closed
2 months ago
Comments count
7
Possible problems about the training dataset
Closed
3 months ago
Comments count
2
Why are the batch size and number of epochs much larger than common SFT settings?
Closed
3 months ago
Comments count
3
Is the model initialized from pre-trained model or model from the last iteration round for each round?
Closed
3 months ago
Comments count
2
Why not directly generate the value, but instead add a value head? Could you explain the reasoning behind this decision?
Closed
4 months ago
Comments count
1
value estimation twice?
Closed
4 months ago
Comments count
5
training code
Closed
4 months ago
Comments count
2
How to initialize first generation child nodes?
Closed
4 months ago
Comments count
1
How to set B1 in Step level Beam Search
Closed
4 months ago
Comments count
3
MCTS training data generation in round1
Closed
4 months ago
Comments count
1
AttributeError: 'RequestOutput' object has no attribute 'value_estimate'
Closed
4 months ago
Comments count
1
About the code
Closed
5 months ago
Comments count
3
数学推理本身是个非对称二元博弈问题
Closed
5 months ago
Comments count
3
AlphaMath listed as AlaphaMath in Huggingface
Closed
5 months ago
Comments count
1
Concern on (first few rounds) sampling efficacy
Closed
5 months ago
Comments count
3