INDEX

Explanations

the verb "are" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 9-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_9/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.9.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

SIN

-0.61

SIZ

-0.61

 Augusta

-0.60

icity

-0.60

žin

-0.60

vänt

-0.59

ilit

-0.58

axel

-0.58

 sjuk

-0.58

 vezet

-0.56

POSITIVE LOGITS

are

1.32

 yourselves

1.19

ARE

1.17

 were

1.13

 themselves

1.07

 WERE

1.06

wolves

1.04

were

1.01

Are

1.01

 Were

1.00

Activations Density 0.478%

the verb "are" and its variations in different contexts

No Comments

No Known Activations