Datasets:

lucasmccabe-lmi
/

openai_humaneval_alpaca_style

Modalities:

Formats:

Size:

Libraries:

Dataset card Viewer Files Files and versions Community

lucasmccabe-lmi commited on Apr 25, 2023

Commit

3eca6af

•

1 Parent(s): c058cd6

Update README.md

Files changed (1) hide show

README.md +5 -1

README.md CHANGED Viewed

@@ -16,4 +16,8 @@ dataset_info:
 ---
 # Dataset Card for "openai_humaneval_alpaca_style"
-[More Information needed](https://github.com/huggingface/datasets/blob/main/CONTRIBUTING.md#how-to-contribute-to-the-dataset-cards)

 ---
 # Dataset Card for "openai_humaneval_alpaca_style"
+This dataset consists of hand-written Python solutions to 164 programming problems by OpenAI. This dataset is equivalent to [openai_humaneval](https://huggingface.co/datasets/openai_humaneval), but has been slightly modified to fit the Alpaca format and include an input field ("The prompt's code is written in Python. Write corresponding response code in Python, as well.") when Python is not explicitly mentioned in the prompt.
+Numbers:
+- **Prompts**: 164
+- **Tokens**: 36644 using the [EleutherAI/gpt-neox-20b](https://huggingface.co/EleutherAI/gpt-neox-20b) tokenizer (counting instruction+input+output)