INDEX

Explanations

Names

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

essim

-0.24

ÐµÑģÑĤÐ¸

-0.23

"group

-0.23

missive

-0.22

igrams

-0.22

NotNil

-0.22

ioneer

-0.22

[number

-0.22

º«

-0.22

osite

-0.22

POSITIVE LOGITS

éĴŁ

0.29

Dil

0.27

Us

0.25

pet

0.24

ãĤĪãģĨ

0.24

å¯¹éĿ¢

0.24

è¿ģ

0.24

med

0.23

bben

0.23

Activations Density 0.019%

Names

No Comments

No Known Activations