© 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

LLAMA3.1-8B · 17-LLAMASCOPE-RES-131K · 39001 ｜ Neuronpedia

Home
OpenMOSS · Llama Scope: SAEs for Llama-3.1-8B
Llama3.1-8B (Base)
Residual Stream
17-LLAMASCOPE-RES-131K
39001

INDEX

Explanations

instances of the word "Correct."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Configuration

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

quier

-0.08

ekim

-0.08

Debe

-0.08

 iÃ§i

-0.08

è¸

-0.08

illez

-0.07

Ø¯Ø§Ø¯

-0.07

iros

-0.07

apÄ±

-0.07

/bower

-0.07

POSITIVE LOGITS

ness

0.07

 Amir

0.06

ula

0.06

poc

0.06

bage

0.06

ly

0.05

CWE

0.05

ity

0.05

ue

0.05

ph

0.05

Activations Density 0.000%

No Known Activations