INDEX

Explanations

cause

np_max-act · gemini-2.0-flash

New Auto-Interp

Configuration

mwhanna/qwen3-4b-transcoders/layer_22.safetensors

Prompts (Dashboard)

16,384 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

163,840

Data Type

float32

Hook Name

blocks.22.mlp.hook_in

Architecture

transcoder

Context Size

8,192

Dataset

monology/pile-uncopyrighted

Embeds

IFrame

Link

Negative Logits

 Aircraft

-0.28

esc

-0.26

æŀľ

-0.25

lege

-0.25

ander

-0.25

alian

-0.25

åĪĹ

-0.25

 contin

-0.24

gin

-0.24

 containing

-0.24

POSITIVE LOGITS

åĨĴ

0.28

éĽ¶è·Ŀç¦»

0.26

 cramped

0.25

èĴĮ

0.24

åĵĳ

0.24

Enums

0.24

enos

0.23

etsy

0.23

)::

0.23

less

0.23

Activations Density 0.009%

cause

No Known Activations