Maksymilian PRO

ucr-max

3 1 7

szorcmax

AI & ML interests

None yet

Recent Activity

updated a Space about 4 hours ago

UniversalComputingResearch/README

repliedto their post about 4 hours ago

Published under Universal Computing Research: Atom2.7m, a 2.74M-parameter causal LM experiment with arithmetic-aware numeric representation. It keeps regular BPE-style text continuation, while adding structure for numeric spans: digit ordering, place_ids, and role_ids. Atom2.7m reaches 69.24% accuracy on ArithMark2.0. The question behind the project is narrow: how much of small-model arithmetic performance depends on representation, rather than scale alone? Model: https://huggingface.co/UniversalComputingResearch/Atom2.7m Article: https://huggingface.co/blog/ucr-max/atom2-7m-arithmetic-representation Demo: https://huggingface.co/spaces/UniversalComputingResearch/Atom2.7-Demo

posted an update about 4 hours ago

View all activity

Organizations

updated a Space about 4 hours ago

Universal Computing Research

🌍

replied to their post about 4 hours ago

Feedback is welcome, especially on useful ablations: digit reversal, place_ids, role_ids, or whether auxiliary features should extend beyond arithmetic spans into ordinary text.

posted an update about 4 hours ago

Post

Published under Universal Computing Research: Atom2.7m, a 2.74M-parameter causal LM experiment with arithmetic-aware numeric representation.

It keeps regular BPE-style text continuation, while adding structure for numeric spans: digit ordering, place_ids, and role_ids.

Atom2.7m reaches 69.24% accuracy on ArithMark2.0.

The question behind the project is narrow: how much of small-model arithmetic performance depends on representation, rather than scale alone?

Model: UniversalComputingResearch/Atom2.7m
Article: https://huggingface.co/blog/ucr-max/atom2-7m-arithmetic-representation
Demo: UniversalComputingResearch/Atom2.7-Demo