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 Space 1 day ago
LabHC/dike-leaderboard liked a Space 1 day ago
iproskurina/compressed_llm_safety_leaderboard updated a Space 1 day ago
iproskurina/compressed_llm_safety_leaderboardOrganizations
Attention graph features extracted from LMs fine-tuned on linguistic acceptability corpora
-
iproskurina/tda-ruroberta-large-ru-cola
Text Classification • 0.4B • Updated • 49 -
iproskurina/tda-bert-en-cola
Text Classification • 0.1B • Updated • 22 -
iproskurina/tda-roberta-large-en-cola
Text Classification • 0.4B • Updated • 30 -
iproskurina/tda-rubert-ru-cola
Text Classification • 0.2B • Updated • 12
A collection of models submitted to the BabyLM23 competition
LLMs quantized with GPTQ
-
iproskurina/Mistral-7B-v0.3-GPTQ-4bit-g128
Text Generation • 7B • Updated • 11 -
iproskurina/bloom-7b1-GPTQ-4bit-g128
Text Generation • 3B • Updated • 38 • 2 -
iproskurina/bloom-1b7-GPTQ-4bit-g128
Text Generation • 1B • Updated • 34 -
iproskurina/bloom-3b-GPTQ-4bit-g128
Text Generation • 2B • Updated • 37
A curated collection of datasets in French, designed to evaluate biases in LLMs
LLMs quantized with GPTQ
-
iproskurina/Mistral-7B-v0.3-GPTQ-4bit-g128
Text Generation • 7B • Updated • 11 -
iproskurina/bloom-7b1-GPTQ-4bit-g128
Text Generation • 3B • Updated • 38 • 2 -
iproskurina/bloom-1b7-GPTQ-4bit-g128
Text Generation • 1B • Updated • 34 -
iproskurina/bloom-3b-GPTQ-4bit-g128
Text Generation • 2B • Updated • 37
Attention graph features extracted from LMs fine-tuned on linguistic acceptability corpora
-
iproskurina/tda-ruroberta-large-ru-cola
Text Classification • 0.4B • Updated • 49 -
iproskurina/tda-bert-en-cola
Text Classification • 0.1B • Updated • 22 -
iproskurina/tda-roberta-large-en-cola
Text Classification • 0.4B • Updated • 30 -
iproskurina/tda-rubert-ru-cola
Text Classification • 0.2B • Updated • 12
A collection of models submitted to the BabyLM23 competition