INDEX

Explanations

phrases indicating presence or existence, particularly in reference to the reader

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Autoritní

-0.86

INGHAM

-0.62

 Roskov

-0.60

 bianchi

-0.58

bueno

-0.57

 blanches

-0.57

 TestBed

-0.56

IUrlHelper

-0.55

medriver

-0.54

 <<<<<<<<<<<<<<

-0.54

POSITIVE LOGITS

You

0.81

 Yous

0.79

ورك

0.76

you

0.75

 yourself

0.72

 محفوظة

0.71

 youre

0.70

You

0.69

Вам

0.69

جوايز

0.68

Activations Density 0.229%

phrases indicating presence or existence, particularly in reference to the reader

No Comments

No Known Activations