This repo is used for LLM Benchmark.
| model | Acc@lambada_openai | Acc@cbt-cn-val | Acc@cbt-cn-test | Acc@cbt-ne-val | Acc@cbt-ne-test |
|---|---|---|---|---|---|
| gpt2 117M | 44.7 | 87.05 | 86.12 | 84.35 | 79.44 |
| gpt2 117M (paper) | 45.99 | 87.65 | * | 83.4 | * |
KeyKy/llm_benchmark
Folders and files
| Name | Name | Last commit date | ||
|---|---|---|---|---|