INDEX

Explanations

ilar

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_23.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.23.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

åĲĦè¡Į

-0.27

çĹĥ

-0.27

 fraction

-0.26

è·³åĩº

-0.26

 Ð½Ð°Ð·Ð°Ð´

-0.25

 sticking

-0.25

 bubble

-0.25

 regular

-0.24

 equivalents

-0.23

WithPath

-0.23

POSITIVE LOGITS

pects

0.31

sei

0.30

een

0.27

çł

0.25

etto

0.25

gin

0.25

icate

0.25

ei

0.24

omal

0.24

ä¸įè§£

0.24

Activations Density 0.106%

ilar

No Comments

No Known Activations