INDEX

Explanations

instances of text that represent structured data or formatted content

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

 propOrder

-0.60

 CreateTagHelper

-0.59

lrrrr

-0.56

endphp

-0.53

 mouseClicked

-0.52

OrBuilder

-0.52

 êtes

-0.52

 TestBed

-0.51

olol

-0.50

 CanadaChoose

-0.49

POSITIVE LOGITS

enterOuterAlt

0.65

ImageContext

0.59

 tyme

0.56

 onsite

0.55

 beech

0.53

 marbles

0.52

nefs

0.51

 NUKAT

0.50

 Jove

0.49

 wellness

0.49

Activations Density 0.005%

instances of text that represent structured data or formatted content

No Comments

No Known Activations