Datasets:

tts-research
/

DeepSeek-R1-Distill-Qwen-1.5B-PRM-prm800k-Llama-3.2-3B-Instruct-best_of_n-completions

Modalities:

Formats:

Size:

Libraries:

Dataset Viewer

Subset (5)

MATH-500--T-0.8--top_p-1.0--n-256--max_tokens-2048--bsz-8--seed-0--agg_strategy-last--evals · 9 rows

Split (1)

train · 9 rows

Search is not available for this dataset

n int64 1 256	acc_naive float64 45.6 55.8	acc_weighted float64 45.6 65.8	acc_maj float64 45.6 64.2
2	51.2	51.2	45.6
32	55.8	62.8	62
16	55	59.6	60.2
256	54.6	65.2	63.8
64	55	64.4	62.6
8	55	58.2	56.8
4	53	54.2	50
1	45.6	45.6	45.6
128	54.6	65.8	64.2

README.md exists but content is empty.