SFT-Mistral-7B (Metamath + ovm +ensemble)

Reported on 2 benchmarks across 1 task

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Reasoning2 results

Arithmetic ReasoningonGSM8K
Accuracy· uses extra data
94.13
best: 97.72 (Claude 3.5 Sonnet (HPT))
Arithmetic ReasoningonGSM8K
Parameters (Billion)· uses extra data
7
best: 540 (PaLM 540B (Self Improvement, Self Consistency))