INDEX

Explanations

Conjunctions and commas

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

 Apprec

-0.29

.sec

-0.26

 appreciate

-0.25

.gallery

-0.25

 sometime

-0.24

+(

-0.24

ØªØµ

-0.24

 Euler

-0.23

som

-0.23

 Secure

-0.23

POSITIVE LOGITS

æĺ¯åĲ¦æľī

0.47

 whether

0.44

æĺ¯åĲ¦

0.43

 Whether

0.42

whether

0.40

æĺ¯åĲ¦åŃĺåľ¨

0.39

Whether

0.38

æĺ¯åĲ¦ä¼ļ

0.36

èĥ½åĲ¦

0.36

 size

0.35

Activations Density 0.004%

Conjunctions and commas

No Comments

No Known Activations