INDEX

Explanations

passive voice phrases indicating actions performed by someone or something in formal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

WillAppear

-0.58

 nahilalakip

-0.54

<bos>

-0.49

ándor

-0.49

hasMoreElements

-0.48

Or

-0.48

omitempty

-0.46

-0.45

Pal

-0.45

PickerController

-0.44

POSITIVE LOGITS

by

0.89

 oleh

0.88

 raiſ

0.84

 uſed

0.81

 becauſe

0.79

 Diretto

0.76

 الفلك

0.74

 itſelf

0.72

 esternos

0.72

ByUser

0.72

Activations Density 0.606%

passive voice phrases indicating actions performed by someone or something in formal contexts

No Comments

No Known Activations