INDEX

Explanations

references to the word "one" and its variations or context in different forms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

BibitemShut

-0.59

 desplaz

-0.56

ecimento

-0.55

 cuisson

-0.53

 fazenda

-0.52

faßt

-0.52

zzleHttp

-0.52

 caminhada

-0.51

 régal

-0.51

 amizade

-0.51

POSITIVE LOGITS

ONE

1.09

one

1.08

One

1.05

One

1.04

one

0.92

HasOne

0.81

ONE

0.79

findOne

0.78

getOne

0.76

 findOne

0.73

Activations Density 0.220%

references to the word "one" and its variations or context in different forms

No Comments

No Known Activations