INDEX

Explanations

Legal or formal language

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_35.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.35.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 overflowing

-0.25

 scrollTop

-0.25

 overflow

-0.25

OBJ

-0.25

CI

-0.25

çľº

-0.24

OT

-0.24

presso

-0.24

 crunch

-0.24

(unique

-0.23

POSITIVE LOGITS

imer

0.38

/im

0.33

 imper

0.32

ramid

0.31

im

0.31

ennen

0.30

jam

0.30

-im

0.30

0.29

jab

0.29

Activations Density 0.012%

Legal or formal language

No Known Activations