INDEX

Explanations

psychology and behavior

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_31.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.31.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

leine

-0.29

ulumi

-0.27

ucer

-0.26

äººåľ¨

-0.26

bett

-0.25

æ·±èĢķ

-0.25

zano

-0.24

 lanc

-0.24

>equals

-0.24

 eens

-0.24

POSITIVE LOGITS

æĪĳä»¬çļĦ

0.36

 ourselves

0.35

our

0.34

OUR

0.30

 ours

0.29

Our

0.29

Our

0.29

æľ¬æĬ¥

0.28

DataRow

0.28

èĮ¹

0.28

Activations Density 0.705%

psychology and behavior

No Known Activations