© 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 · 26139 ｜ Neuronpedia

Home
Olmo-3-1125-32B
32-RES-BATCHTOPK-131K
26139

INDEX

Explanations

of

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

iton

-0.09

izzare

-0.08

Attr

-0.08

 oppression

-0.08

ima

-0.08

ucci

-0.08

ogg

-0.08

extension

-0.08

 Orig

-0.08

ux

-0.08

POSITIVE LOGITS

ably

0.11

hap

0.11

how

0.11

 impending

0.10

alus

0.10

 dire

0.09

0.09

 goose

0.09

0.09

ordin

0.09

Activations Density 0.049%

No Known Activations