INDEX

Explanations

Institutional contexts

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

å®Į

-0.27

HP

-0.26

 fraction

-0.26

 fractions

-0.25

 edge

-0.25

æłı

-0.25

åįĩ

-0.25

èĥ«

-0.24

çŁ¥

-0.24

orry

-0.24

POSITIVE LOGITS

 complementary

0.26

¦Ĥ

0.26

 competing

0.26

gradable

0.25

panion

0.25

MimeType

0.24

wers

0.24

åłĥ

0.24

upid

0.23

(Equal

0.23

Activations Density 0.004%

Institutional contexts

No Comments

No Known Activations