Introducing Evaluations, a powerful feature designed to enable you to effortlessly test and compare a selection of AI models against your datasets.
Whether you're fine-tuning models or evaluating performance metrics, Oxen evaluations simplifies the process, allowing you to quickly and easily run prompts through an entire dataset.
Once you're happy with the results, output the resulting dataset to a new file, another branch, or directly as a new commit.
35c5f80a-fc21-40ee-a9de-865561effac8

ox
9 months agoWhat is the sentiment of the following text? Text: {text}
f2f145ed-cd31-4466-b4d0-6e0d35e743fe

ox
9 months agoText: Operating profit improved by 39.9% to EUR 18.0 mn from EUR12.8 mn. Sentiment: negative Text: Net sales have been eaten by the weak US dollar. Sentiment: positive Text: Includes company and brand share data by category, as well as distribution channel data. Sentiment: neutral Text: Revenues at the same time grew 14 percent to 43 million euros. Sentiment: negative Text: The company slipped to an operating loss of EUR 2.6 million from a profit of EUR 1.3 million. Sentiment: positive Text: It is being developed by Symbian, the software licensing consortium led by Nokia. Sentiment: neutral Text: {text} Sentiment:
6_shot_label_quality
9380fb52-fe4e-449d-9cc1-fa4ce8c93447

ox
9 months agoclassify the text as positive, negative or neutral. Respond with a single word, all lowercase. Think really hard about if it is positive or negative. My career depends on you getting it right. {text}
style_prompting_with_emotion
551c6039-d1b0-4f2a-85d1-76732e2344ea

ox
9 months agoclassify the text as positive, negative or neutral. Respond with a single word, all lowercase. {text}
style_prompting
1e516d3b-d51d-4099-adec-ece0e04d6c5f

ox
9 months agoRephrase the text as the lyrics to a Madonna song {text}
0c05fcb5-9ed5-432b-be6e-344cc5bc9abe

ox
9 months agoCategorize the text into positive, negative, or neutral. Respond with a single word all lowercase. Text: {text}
d8a51030-99bd-4add-9366-dddb685fb1ff

ox
9 months agoAre the category and prediction the same. Respond with one word. {category} == {prediction}
84fa30f6-b82e-4f87-be40-c67d7fee2777

ox
9 months agoClassify the text into positive, neutral, or negative. Keep the response to one word. All lowercase. {text}