INDEX

Explanations

commas

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

ivals

-0.29

çľĭä¸Ģçľĭ

-0.29

 çĻ»

-0.26

-REAL

-0.25

gettext

-0.24

.timedelta

-0.24

åįģéĩĮ

-0.24

 panicked

-0.23

ardin

-0.23

COMP

-0.23

POSITIVE LOGITS

omer

0.35

ãģ®æĸ¹

0.27

urus

0.25

æĭ¼

0.25

çķĻä¸ĭäºĨ

0.25

 Plate

0.24

Plate

0.24

mos

0.24

in

0.23

åĳ½ä¸Ń

0.23

Activations Density 0.027%

commas

No Comments

No Known Activations