INDEX

Explanations

references to navigation and user interface elements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.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

TargetException

-0.66

 pulumi

-0.63

ècie

-0.54

ifornie

-0.54

ändigkeit

-0.52

 geografic

-0.52

 ContentValues

-0.52

LEEP

-0.51

])));

-0.50

esModule

-0.49

POSITIVE LOGITS

 AppBar

0.71

menuStrip

0.62

cabec

0.59

 Navbar

0.57

BeginInit

0.57

cdef

0.56

toolBar

0.55

RetentionPolicy

0.54

 actionBar

0.53

 navbar

0.53

Activations Density 0.031%

references to navigation and user interface elements

No Comments

No Known Activations