Arch Gale
In this project, we design prompts based on predefined categories with the specific goal of testing the model's limitations and prompting it to fail.
Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
No subject matter listed
No task types listed
In my recent role as a Data Annotator at Outlier, I’ve built strong expertise in AI training data, completing over 4,000 tasks in the past nine months. I’ve worked as both an attempter, creating initial labels, and a reviewer, ensuring top data quality and guideline alignment. This dual experience gives me a full view of the annotation process and enables me to consistently deliver accurate, high-quality datasets.
In this project, we design prompts based on predefined categories with the specific goal of testing the model's limitations and prompting it to fail.
In this project we evaluate prompts and model responses against criteria such as factuality, instruction following, tone, formatting, etc.
In this project we evaluate the response against 8 dimensions: Accuracy, Instruction Following, Refusal Correctness, Relevance, Comprehensiveness, Relevance, Grammar & Presentation, and Tone/Style.
Bachelor in Communication Sciences, Communication Sciences
Generalist