INDEX

Explanations

structures related to unique identifiers or categories, particularly variations of the word "one."

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

 насељу

-0.89

 ―――――

-0.79

hésite

-0.71

Cordialement

-0.71

CrossRef

-0.70

apunov

-0.70

bcryptjs

-0.69

 ſch

-0.68

 SearchView

-0.68

ccm

-0.67

POSITIVE LOGITS

One

1.93

One

1.91

ONE

1.84

one

1.82

one

1.81

ONE

1.69

getOne

1.18

 onep

1.17

 Ones

1.14

 één

1.13

Activations Density 0.151%

structures related to unique identifiers or categories, particularly variations of the word "one."

No Comments

No Known Activations