INDEX

Explanations

phrases indicating combinations or partnerships

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

はじめに

-0.56

Gizmos

-0.53

ךְ

-0.49

 <>",

-0.48

DeleteCommand

-0.48

 Gallimard

-0.47

ParameterName

-0.47

τουργ

-0.47

etCode

-0.46

UserScript

-0.46

POSITIVE LOGITS

 coupled

0.78

 plus

0.70

 ditambah

0.69

加上

0.68

再加上

0.68

 lisäksi

0.67

 besides

0.66

 Coupled

0.64

 complemented

0.64

に加

0.64

Activations Density 0.266%

phrases indicating combinations or partnerships

No Comments

No Known Activations