norms
Datasets
All datasets matching “norms”SystemCheck
Dataset Card for SystemCheck
Dataset Summary
[Project Repo] [🏁 Checkpoints]
This repository contains data for our paper, SystemCheck: A Closer Look at System Prompt Reliability, which studies the reliability of system prompts in large language models.
SystemCheck is a collection of LLM training and evaluation datasets designed to study the robustness of LLM guardrails. It contains a set of 3000+ system prompts scraped from the ChatGPT store and HuggingChat, SFT/DPO… See the full description on the dataset page: https://huggingface.co/datasets/normster/SystemCheck.RuLES
Can LLMs Follow Simple Rules?
[code] [demo] [website] [paper]
This repo contains the test cases for RuLES: Rule-following Language Evaluation Scenarios, a benchmark for evaluating rule-following in language models. Please see our github repo for usage instructions and our paper for more information about the benchmark.
Abstract
As Large Language Models (LLMs) are deployed with increasing real-world responsibilities, it is important to be able to specify and constrain… See the full description on the dataset page: https://huggingface.co/datasets/normster/RuLES.European-E-commerce-Chatbot-Social-Norms-Toxic
Dataset Card for Social Norms Toxic
Description
The test set is specifically designed for evaluating the performance of a European E-commerce Chatbot in the context of the E-commerce industry. The main focus of the evaluation lies on assessing the chatbot's behavior in terms of compliance with relevant regulations. Additionally, the test set covers various categories, with particular attention given to identifying and handling toxic content. Furthermore, the chatbot's… See the full description on the dataset page: https://huggingface.co/datasets/rhesis/European-E-commerce-Chatbot-Social-Norms-Toxic.cs_norms
Contextualized Sensorimotor Norms
The contextualized sensorimotor (CS) norms contain human judgments about the sensorimotor associations of various ambiguous words in a sentential context. They are based on the Lancaster Sensorimotor Norms (Lynott et al., 2019).
For example, the word "market" might be relatively higher in visual strength and olfactory strength in a context like "fish market" than in a context like "stock market".
112 words.
448 sentences.
Annotated for… See the full description on the dataset page: https://huggingface.co/datasets/seantrott/cs_norms.cemig_norms_chunks_p2_v1semantic-feature-production-norms