INDEX

Explanations

instances of negation or absence expressed through words like "no" and "none."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

 joaat

-0.63

gartner

-0.51

 Wilton

-0.50

fromnode

-0.49

Pastor

-0.49

Loren

-0.48

 thiệu

-0.48

DataAnnotations

-0.48

adays

-0.47

gino

-0.45

POSITIVE LOGITS

LookAnd

0.78

Nowhere

0.73

utilisons

0.72

 neither

0.69

InjectAttribute

0.69

neither

0.67

postsleuth

0.66

 Stair

0.64

tram

0.61

 définiti

0.61

Activations Density 0.166%

instances of negation or absence expressed through words like "no" and "none."

No Comments

No Known Activations