© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals Exports interp-engineNEWAPI Community Blog Privacy & Terms Contact

OLMO-3-1125-32B · 32-RES-BATCHTOPK-131K · 92252 ｜ Neuronpedia

Home
Olmo-3-1125-32B
32-RES-BATCHTOPK-131K
92252

INDEX

Explanations

finding a way to

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Configuration

bcywinski/Olmo-3-32B-Base-SAE/saes_allenai_Olmo-3-1125-32B_batch_top_k/resid_post_layer_32

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

lf

-0.10

ussen

-0.10

 Amber

-0.10

cola

-0.10

oris

-0.10

 deciding

-0.10

ceso

-0.09

 Alta

-0.09

defs

-0.09

ressing

-0.09

POSITIVE LOGITS

 somehow

0.20

 somew

0.18

way

0.15

 Somehow

0.15

 ways

0.14

 workaround

0.14

way

0.10

 bypass

0.10

 looph

0.10

Zwe

0.10

Activations Density 0.053%

No Known Activations