INDEX

Explanations

Code/automatically generated content

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

::__

-0.28

olars

-0.27

 silenced

-0.26

(Seq

-0.26

_pll

-0.25

-route

-0.25

ulings

-0.25

itto

-0.24

éĿ¢åīį

-0.24

loggedin

-0.24

POSITIVE LOGITS

æĬĬå®ĥ

0.27

ization

0.27

å¹»

0.26

è®°å½ķ

0.26

ORT

0.25

ty

0.25

umes

0.25

 record

0.25

èĦ±

0.25

åı¦ä¸Ģç§į

0.25

Activations Density 0.002%

Code/automatically generated content

No Comments

No Known Activations