INDEX

Explanations

the beginning of new sections or statements

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

 useContext

-0.68

SharedDtor

-0.64

expandindo

-0.56

 referenties

-0.54

 collectively

-0.52

 bane

-0.51

ValueStyle

-0.50

 виправивши

-0.50

oredCriteria

-0.49

 something

-0.48

POSITIVE LOGITS

 الحره

0.94

scrapy

0.57

AppMethodBeat

0.56

 inggris

0.54

scalatest

0.54

)|^{

0.53

isissez

0.53

 pertanto

0.52

ifrance

0.52

DOCTYPE

0.52

Activations Density 0.014%

the beginning of new sections or statements

No Comments

No Known Activations