Dan P

Datdanboi25's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in Glint-Research/Glint-1.3 about 3 hours ago

scaling laws

👍 1

13

#2 opened 24 days ago by

encryptedoreo

liked 2 models about 5 hours ago

SupraLabs/cma-8M

Text Generation • 9.08M • Updated about 2 hours ago • 5

PhysiQuanty/arith-test

856k • Updated about 9 hours ago • 2

updated a Space about 17 hours ago

Running

58

Open SLM Leaderboard

🏆

58

Open Small Language Model Leaderboard

liked a model about 24 hours ago

MaliosDark/Nexus-Erebus-3M

Text Generation • 3M • Updated 2 days ago • 558 • 3

New activity in AxiomicLabs/Open_SLM_Leaderboard 1 day ago

Add Nexus-Erebus-3M (from-scratch 3.0M, arithmetic specialist)

1

#40 opened 2 days ago by

MaliosDark

liked 2 models 2 days ago

AxionLab-official/RizzAura-100k

Text Generation • Updated 2 days ago • 198 • 3

Nikity/lille-130m-base

Text Generation • 0.1B • Updated Sep 4, 2025 • 88 • 16

liked a Space 3 days ago

Running on Zero

Agents

2

Gros Michel Translator

🦀

2

A space to try out the Gros Michel Translation model.

liked a model 3 days ago

syntropic-clx/Clyx_0.2-115.67M-BASE

Text Generation • 0.1B • Updated Jun 7 • 42 • 3

New activity in AxiomicLabs/Open_SLM_Leaderboard 4 days ago

Add Nexus-Erebus-50M (Ideoa Labs, from-scratch 54M)

2

#39 opened 4 days ago by

MaliosDark

liked a model 4 days ago

MaliosDark/Nexus-Erebus-50M

Text Generation • 54.1M • Updated 4 days ago • 655 • 4

New activity in AxiomicLabs/Open_SLM_Leaderboard 5 days ago

Add https://huggingface.co/BananaMind/BananaMind-2-Mini

1

#37 opened 5 days ago by

Banaxi-Tech

Hierarchos experiment

3

#38 opened 5 days ago by

netcat420

liked 3 models 5 days ago

netcat420/Hierarchos-experiment

Updated Jan 21 • 1

BananaMind/BananaMind-2-Mini

Text Generation • 28.3M • Updated 5 days ago • 852 • 9

netcat420/KortexHOS

Text Generation • Updated 7 days ago • 2

reacted to ucr-max's post with 🔥 5 days ago

view post

Post

102

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

See translation

• 

• 
• 1 reply

·

🔥

4

4

+

Reply

New activity in AxiomicLabs/Open_SLM_Leaderboard 6 days ago

Add Banaxi-Tech/waterfall-alpha

6

#36 opened 7 days ago by

Banaxi-Tech

upvoted an article 6 days ago

view article

Article

Atom2.7m: Representation-Level Specialization for Arithmetic-Aware Small Language Models

• 

ucr-max

•

6 days ago

• 9