INDEX

Explanations

definite articles indicating importance or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

ритори

-0.69

 Biôgrafia

-0.68

 <>",

-0.64

SourceChecksum

-0.64

 Roskov

-0.64

 betweenstory

-0.62

 GenerationType

-0.61

こち

-0.59

LogFactory

-0.58

 resourceCulture

-0.57

POSITIVE LOGITS

key

1.40

 best

1.20

 thing

1.18

 easiest

1.14

 important

1.11

 more

1.10

 good

1.10

 trick

1.06

 reason

1.05

 problem

1.03

Activations Density 0.343%

definite articles indicating importance or emphasis

No Comments

No Known Activations