INDEX

Explanations

Naming conventions

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_7.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.7.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

uality

-0.28

xit

-0.27

reak

-0.27

udent

-0.26

 Spart

-0.25

 contro

-0.25

-handed

-0.25

ÑħÐ¾Ð´

-0.25

 nations

-0.25

inus

-0.24

POSITIVE LOGITS

Tu

0.31

Tu

0.26

osa

0.26

Svc

0.25

slideUp

0.25

åľ¨åĮĹäº¬

0.24

å¼¼

0.24

æħ¢

0.24

iloc

0.24

rams

0.24

Activations Density 0.001%

Naming conventions

No Comments

No Known Activations