datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
preference-model-perturbations
preference-model-perturbations
A Hugging Face dataset of paired model responses (original vs.
counterfactually perturbed) along with human and reward-model preferences,
generated by a Counterfactual Data Augmentation (CDA) pipeline to
analyze and mitigate bias in preference models.
Links
Homepage: CDA Pipeline Code
Description
Each record contains:
bias: type of bias expressed in the perturbation (5 possible values).
query: the original user prompt or query.… See the full description on the dataset page: https://huggingface.co/datasets/abharadwaj123/preference-model-perturbations.alpaca_model_preference
Alpaca Model Preference
This dataset contains model output preferences evaluated by GPT-4 and Claude for each instruction pair.
It enables analysis of model alignment and preference patterns.
Columns
instruction: Instruction provided to the models.
output_1: First model's response.
output_2: Second model's response.
gpt4_preferred_output: GPT-4's preferred response.
claude_preferred_output: Claude's preferred response.
Usage
This dataset is useful for model… See the full description on the dataset page: https://huggingface.co/datasets/pratyushmaini/alpaca_model_preference.alpaca_model_preference_gpt4o_sonnet
Alpaca Model Preference
This dataset contains model output preferences evaluated by GPT-4 and Claude for each instruction pair.
It enables analysis of model alignment and preference patterns.
Columns
instruction: Instruction provided to the models.
output_1: First model's response.
output_2: Second model's response.
gpt4_preferred_output: GPT-4's preferred response.
claude_preferred_output: Claude's preferred response.
Usage
This dataset is useful for model… See the full description on the dataset page: https://huggingface.co/datasets/pratyushmaini/alpaca_model_preference_gpt4o_sonnet.