GPT-2-Medium 355M (BS=5)

Reported on 2 benchmarks across 1 task · 1 paper

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

Reasoning2 results

Arithmetic ReasoningonGSM8K
Accuracy· 2022-10-20
12.2
best: 97.72 (Claude 3.5 Sonnet (HPT))
Composing Ensembles of Pre-trained Models via Iterative Consensus arXiv:2210.11522
Arithmetic ReasoningonGSM8K
Parameters (Billion)· 2022-10-20
0.355
best: 540 (PaLM 540B (Self Improvement, Self Consistency))
Composing Ensembles of Pre-trained Models via Iterative Consensus arXiv:2210.11522