INDEX

Explanations

special formatting or structure indicative of programming or technical content

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

code snippets involving the declaration of functions and methods

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

Self health management

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

writeFieldEnd

-0.94

 дописавши

-0.92

 Wiktionnaire

-0.83

 فريبيس

-0.81

 Theſe

-0.79

Przypisy

-0.78

Приятного

-0.76

 AssemblyCulture

-0.75

 censiti

-0.73

 tartalomajánló

-0.73

POSITIVE LOGITS

0.47

“

0.47

’

0.46

·

0.45

办

0.45

ma

0.44

ve

0.44

is

0.44

 IllegalArgument

0.43

“

0.43

Activations Density 0.012%

special formatting or structure indicative of programming or technical content

code snippets involving the declaration of functions and methods

Self health management

No Comments

No Known Activations