INDEX

Explanations

terms related to image processing and dataset analysis

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

outdoor scene illumination changes

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 informée

-0.33

 확인

-0.32

 hizmet

-0.31

 angeb

-0.31

BeginContext

-0.30

 spos

-0.29

 insolvency

-0.28

 supuestamente

-0.28

ByVersion

-0.28

 Hayden

-0.27

POSITIVE LOGITS

 lenker

0.64

Rüyada

0.59

 ویکی‌پدیا

0.55

AISSEE

0.53

 nahilalakip

0.52

 незавершена

0.52

 ddelweddau

0.51

舺

0.51

 cherchés

0.50

ciech

0.49

Activations Density 0.062%

terms related to image processing and dataset analysis

outdoor scene illumination changes

No Comments

No Known Activations