metadata
license: mit
datasets:
- zwhe99/DeepMath-103K
language:
- en
metrics:
- accuracy
base_model:
- zwhe99/Qwen2.5-7B-orz
tags:
- math
- reasoning
- rl
model-index:
- name: DeepMath-Zero-7B
results:
- task:
type: text-generation
dataset:
type: MATH500
name: MATH500
metrics:
- name: pass@1
type: pass@1
value: 0.855
verified: false
- task:
type: text-generation
dataset:
type: AMC23
name: AMC23
metrics:
- name: pass@1
type: pass@1
value: 0.647
verified: false
- task:
type: text-generation
dataset:
type: OlympiadBench
name: OlympiadBench
metrics:
- name: pass@1
type: pass@1
value: 0.51
verified: false
- task:
type: text-generation
dataset:
type: MinervaMath
name: MinervaMath
metrics:
- name: pass@1
type: pass@1
value: 0.453
verified: false
- task:
type: text-generation
dataset:
type: AIME24
name: AIME24
metrics:
- name: pass@1
type: pass@1
value: 0.204
verified: false
- task:
type: text-generation
dataset:
type: AIME24
name: AIME24
metrics:
- name: pass@1
type: pass@1
value: 0.175
verified: false