INDEX

Explanations

phrases related to close-up imagery and zooming in on subjects

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

close-up views

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

ंदीखरीदारी

-0.65

tvguidetime

-0.58

 parachoque

-0.54

pulseira

-0.52

 ícones

-0.49

 للمعارف

-0.49

 transférez

-0.48

 présidentielle

-0.47

pecabe

-0.47

новниш

-0.47

POSITIVE LOGITS

 closeup

0.60

focus

0.50

closeup

0.50

focused

0.45

 CLOSE

0.45

 zoomed

0.45

 detail

0.45

 micros

0.44

 micro

0.44

micro

0.43

Activations Density 0.019%

phrases related to close-up imagery and zooming in on subjects

close-up views

No Comments

No Known Activations