INDEX

Explanations

specifications related to screen size and display features of devices

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

units of measurement

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

Negative Logits

 CreateTagHelper

-0.70

DrawerToggle

-0.54

 consecuencia

-0.54

enumi

-0.53

Aiheesta

-0.52

Literatura

-0.51

 correctes

-0.51

//

-0.50

первых

-0.50

msgTypes

-0.49

POSITIVE LOGITS

TagMode

0.65

IntoConstraints

0.64



0.59

짜

0.58

 jadx

0.56

以内

0.56

Kaynakça

0.55

itre

0.55

cyon

0.54

度的

0.54

Activations Density 0.103%

specifications related to screen size and display features of devices

units of measurement

No Known Activations