© 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

LLAMA3.3-70B-IT · 50-RESID-POST-GF · 36327 ｜ Neuronpedia

Home
Llama3.3-70B-IT
50-RESID-POST-GF
36327

INDEX

Explanations

control and takeover

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.3-70B-IT @ 50-resid-post-gf

Configuration

Goodfire/Llama-3.3-70B-Instruct-SAE-l50/Llama-3.3-70B-Instruct-SAE-l50.pt

Prompts (Dashboard)

10,000 prompts, 128 tokens each

Dataset (Dashboard)

lmsys/lmsys-chat-1m

No Configuration Found

Embeds

Show PlotsShow ExplanationShow ActivationsShow Test FieldShow SteerShow Link

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 annihil

-0.13

mÃŃt

-0.10

 salv

-0.10

apel

-0.10

 trump

-0.09

 Cinema

-0.09

reb

-0.09

abbit

-0.09

Gear

-0.09

oog

-0.09

POSITIVE LOGITS

 control

0.17

 rule

0.15

rid

0.12

-rule

0.12

rule

0.12

 plung

0.12

 plunge

0.12

 Control

0.11

 power

0.11

 gain

0.11

Activations Density 0.057%

No Known Activations