Prompt+Response creation and evaluation
Creation of a localized prompt and evaluation of the two responses provided by the model; selecting the preferred one and rewriting if necessary to ensure it is perfect.
Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
No subject matter listed
I contribute to projects using my expertise in the Italian language. I work both as a contributor and as a reviewer on different kinds of tasks, from evaluation and preference ranking to the creation of prompts and responses. I also have experience in data collection, generating text from speech, and performing safety checks to ensure that the model is harmless. In the past two months, I’ve also been working on projects that include prompts with images: I create them, evaluate the responses, and assess their level of toxicity. I worked on a project where we trained the model to impersonate a character and tested how well it managed to stay in character. In some projects, I’ve become a reviewer of other contributors’ work, and the final line of quality control evaluates my contributions.
Creation of a localized prompt and evaluation of the two responses provided by the model; selecting the preferred one and rewriting if necessary to ensure it is perfect.
Evaluation and preference ranking between two model responses based on accuracy, localization, language quality, instruction following, and truthfulness.
Bachelor, Educational Sciences
Social worker
Front Desk Receptionist