INDEX

Explanations

sections of text that indicate the start of a new idea or topic

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

the start of a document or the beginning of a new section

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

names and identifiers

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

Not in Any Lists

No Comments

Negative Logits

تقاوى

-0.71

()].

-0.69

 ModelExpression

-0.66

}.

-0.66

り

-0.65

&.

-0.62

».

-0.61

+.

-0.60

 insuffisamment

-0.59

坞

-0.59

POSITIVE LOGITS

reportWebVitals

0.64

 وتسجيلات

0.63

NgModule

0.62

ViewInit

0.61

 contextLoads

0.59

ServletContext

0.58

ệc

0.57

 bezeichneter

0.57

#+#

0.56

enumi

0.56

Activations Density 0.024%

sections of text that indicate the start of a new idea or topic

the start of a document or the beginning of a new section

names and identifiers

No Comments

No Known Activations