© Neuronpedia 2026

Privacy & Terms Blog GitHub Slack Twitter Contact

Neuronpedia

Jacobian LensNEW

Natural Language

NEW Assistant AxisNEW Circuit TracerUPDATESteer SAE Evals ExportsAPI Community Blog Privacy & Terms Contact

OLMO-3-1125-32B · 32-RES-BATCHTOPK-131K · 16546 ｜ Neuronpedia

Home
Olmo-3-1125-32B
32-RES-BATCHTOPK-131K
16546

INDEX

Explanations

that's why

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

 infer

-0.12

 sack

-0.11

usty

-0.11

chwitz

-0.11

 guessed

-0.11

 Independ

-0.10

esi

-0.10

<=>

-0.10

Cit

-0.10

.ma

-0.10

POSITIVE LOGITS

why

0.18

 because

0.10

why

0.10

ukan

0.09

 called

0.09

because

0.09

how

0.08

ils

0.08

ture

0.08

 Part

0.08

Activations Density 0.050%

No Known Activations