Small language-model ablations trained on FineWeb-Edu, with evaluation confidence intervals and TensorBoard logs.
Dmitry Balobin PRO
d0rj
P(doom) <0.001%
AI & ML interests
NLP and 🥴 tensors.
MIPT PhD student 💙
Recent Activity
liked a model 5 days ago
yresearch/DMax-Coder-MMD liked a model 6 days ago
codelion/SmolLM2-70M liked a dataset 6 days ago
codelion/sutra-10BOrganizations
Summarization models
Vision datasets
Translated into Russian
Instruct datasets in Russian
All datasets have been translated using Google Translate
Instruct preference datasets in Russian
Collection of Russian instruct data with preferences
Tiny llm ablation
Small language-model ablations trained on FineWeb-Edu, with evaluation confidence intervals and TensorBoard logs.
t5gemma-2 speculators
Summarization models
Retrievers
Vision datasets
Translated into Russian
CoT translated or adapted
into russian
Instruct datasets in Russian
All datasets have been translated using Google Translate
Math Instruct datasets in Russian
Collected Olympiad and final exam tasks, automatically generated tasks, as well as automatically translated datasets. Work in progress...
Instruct preference datasets in Russian
Collection of Russian instruct data with preferences