INDEX

Explanations

Damage and disease

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

Lights

-0.28

hsi

-0.26

 scheme

-0.26

 decking

-0.25

oret

-0.25

Known

-0.24

 dumping

-0.24

äºĭãĤĴ

-0.24

åıįä¹ĭ

-0.24

HORT

-0.23

POSITIVE LOGITS

 part

0.30

ness

0.29

enburg

0.28

è¿Ļéĥ¨åĪĨ

0.28

 areas

0.27

 parts

0.27

è¶ĭ

0.26

ity

0.26

arching

0.25

/Instruction

0.25

Activations Density 0.009%

Damage and disease

No Comments

No Known Activations