INDEX

Explanations

references to religious concepts, particularly those associated with holiness

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

Text following "Holy"

np_acts-logits-general · gemini-2.0-flash

"Holy" followed by common phrases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 незавершена

-1.04

RectangleBorder

-0.94

AccessorTable

-0.94

 שוליים

-0.89

ViewFeatures

-0.89

 صوتيه

-0.87

 المعيارى

-0.83

 itſelf

-0.83

 dAtA

-0.82

setopt

-0.82

POSITIVE LOGITS

 Holy

0.70

Holy

0.67

 holy

0.64

HOLY

0.62

 shit

0.59

 HOLY

0.56

Sh

0.54

mol

0.54

 Ground

0.53

0.52

Activations Density 0.066%

references to religious concepts, particularly those associated with holiness

Text following "Holy"

"Holy" followed by common phrases

No Comments

No Known Activations