INDEX

Explanations

web addresses

np_token-act-pair-logits · gpt-4o-mini

web addresses

oai_token-act-pair · gemini-2.0-flash

www

np_max-act-logits · gemini-2.0-flash

New Auto-Interp

Configuration

google/gemma-scope-2b-pt-transcoders/layer_21/width_16k/average_l0_13

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.ln2.hook_normalized

Architecture

jumprelu_transcoder

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

favicon

-0.49

 AssemblyCompany

-0.48

ByPrimaryKey

-0.47

ContentAsync

-0.46

 ©️

-0.45

ClientSize

-0.45

 https

-0.44

 http

-0.44

deleteById

-0.44

stateParams

-0.44

POSITIVE LOGITS

kegaard

0.69

LLocation

0.60

rrggbb

0.60

 endwhile

0.58

ksikon

0.57

.)}

0.56

 nostru

0.56

potranspiration

0.55

chengladbach

0.55

lccn

0.55

Activations Density 0.061%

web addresses

web addresses

www

No Comments

No Known Activations