INDEX

Explanations

"Gros" or name fragments

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_1.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.1.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

 marking

-0.26

scape

-0.25

ocracy

-0.25

é£Ļ

-0.24

DK

-0.24

ioni

-0.24

éĨĴæĿ¥

-0.24

å°ıæĹ¶åīį

-0.24

ESCO

-0.24

 stains

-0.23

POSITIVE LOGITS

 present

0.29

tet

0.26

 inform

0.26

æķĻ

0.25

 deposit

0.25

Present

0.25

ãģ»ãģ©

0.25

tet

0.25

hom

0.25

Activations Density 33.760%

"Gros" or name fragments

No Comments

No Known Activations