INDEX

Explanations

references to possessive forms, specifically the word "its"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

by

-0.63

er

-0.60

ity

-0.59

to

-0.57

the

-0.55

for

-0.55

had

-0.54

is

-0.51

To

-0.51

-0.50

POSITIVE LOGITS

its

1.87

Its

1.42

Its

1.36

Autoritní

1.21

NameInMap

1.20

它的

1.15

 Савезне

1.08

 дописавши

1.08

 betweenstory

1.01

expandindo

1.01

Activations Density 0.119%

references to possessive forms, specifically the word "its"

No Comments

No Known Activations