A curated collection of datasets in French, designed to evaluate biases in LLMs
Irina Proskurina PRO
iproskurina
AI & ML interests
LLMs: quantization, pre-training
Recent Activity
updated
a dataset 11 minutes ago
iproskurina/bias-in-bios-qwen-hf-r25-iter-new-iter1 published
a dataset 11 minutes ago
iproskurina/bias-in-bios-qwen-hf-r25-iter-new-iter1 published
a model 11 minutes ago
iproskurina/bias-in-bios-qwen-hf-r25-iter-new-iter1 Organizations
Attention graph features extracted from LMs fine-tuned on linguistic acceptability corpora
A collection of models submitted to the BabyLM23 competition
