© 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

LLAMA3.3-70B-IT · 50-RESID-POST-GF · 31665 ｜ Neuronpedia

Home
Llama3.3-70B-IT
50-RESID-POST-GF
31665

INDEX

Explanations

three or numbers

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

 Both

-0.18

Both

-0.16

 ambos

-0.15

ä¸¡

-0.14

duo

-0.14

 Ð¾Ð±Ð¾

-0.14

 obou

-0.14

 BOTH

-0.14

 beide

-0.14

both

-0.14

POSITIVE LOGITS

 three

0.73

 trio

0.61

ä¸īä¸ª

0.61

three

0.59

 THREE

0.54

ä¸ī

0.53

 drei

0.52

 Three

0.52

Three

0.50

-three

0.50

Activations Density 0.487%

No Known Activations