INDEX

Explanations

references to the desires or needs of individuals

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Pronouns followed by "want" or "need"

np_acts-logits-general · gemini-2.0-flash

wants and needs

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Pool

-0.49

 pool

-0.48

 surprised

-0.48

pool

-0.46

met

-0.43

ய

-0.42

tid

-0.42

gate

-0.42

Impression

-0.41

amaz

-0.41

POSITIVE LOGITS

PreferredItem

0.71

 الرياضيه

0.67

 crave

0.66

 pinulongan

0.65

webElementGuid

0.64

SuppressMessage

0.63

Lähteet

0.61

$_[

0.60

waitKey

0.59

대로

0.58

Activations Density 0.120%

references to the desires or needs of individuals

Pronouns followed by "want" or "need"

wants and needs

No Comments

No Known Activations