INDEX

Explanations

code

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

 collabor

-0.27

aÅĤ

-0.24

(al

-0.24

:F

-0.24

shr

-0.24

 Formatting

-0.24

 ally

-0.23

 datatable

-0.23

>Returns

-0.23

Formatting

-0.22

POSITIVE LOGITS

uent

0.27

ä»£çĲĨ

0.25

bring

0.25

Ã»t

0.24

lus

0.24

å¹¹

0.23

/gif

0.23

éģĵ

0.23

 prove

0.23

å½°

0.23

Activations Density 0.001%

code

No Comments

No Known Activations