INDEX

Explanations

phrases or questions that seek to identify or inquire about individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

UnusedPrivate

-0.76

KommentareTeilen

-0.73

 Meksiku

-0.71

adpleegd

-0.68

Javadoc

-0.65

CrossRef

-0.65

 OFDb

-0.65

CloseOperation

-0.64

 تضيفلها

-0.64

PreferredItem

-0.62

POSITIVE LOGITS

Who

0.80

Who

0.79

 else

0.73

who

0.66

who

0.66

WHO

0.58

 antaranya

0.57

Hvem

0.57

 hvem

0.56

 quién

0.56

Activations Density 0.065%

phrases or questions that seek to identify or inquire about individuals

No Comments

No Known Activations