INDEX

Explanations

phrases related to images and visual content

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

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

FormTagHelper

-0.59

AndEndTag

-0.57

 BorderRadius

-0.54

antası

-0.54

 internationaux

-0.54

 InputDecoration

-0.51

UnsafeEnabled

-0.50

bootstrapcdn

-0.48

iastes

-0.48

 hjemme

-0.47

POSITIVE LOGITS

 nakalista

0.66

انيف

0.65

COLLE

0.59

 tartalomajánló

0.58

:✨

0.56

tvguidetime

0.56

 gyhoeddwyd

0.55

)");

0.54

')")

0.54

 مرئيه

0.53

Activations Density 0.186%

phrases related to images and visual content

No Comments

No Known Activations