INDEX

Explanations

programming-related keywords and language-specific identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 noqa

-0.72

 BorderRadius

-0.64

 TestBed

-0.62

TestingModule

-0.61

 TextAppearance

-0.60

forChild

-0.60

aarrggbb

-0.59

featureID

-0.59

ChromeDriver

-0.59

 internetowa

-0.57

POSITIVE LOGITS

IsContent

0.78

APORE

0.69

 مُعرِّف

0.60

ंदीखरीदारी

0.56

uslar

0.56

 AssemblyCulture

0.56

HasForeignKey

0.56

itoneal

0.55

XtraGrid

0.55

Tikang

0.54

Activations Density 0.019%

programming-related keywords and language-specific identifiers

No Comments

No Known Activations