INDEX

Explanations

the usage of the word "we" and variations of it to highlight collective identity or group involvement

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

Chy

-0.68

Chy

-0.63

 Thon

-0.60

хів

-0.56

 fior

-0.56

 funcion

-0.55

WithIOException

-0.55

 açıklama

-0.54

 maris

-0.54

 Moors

-0.53

POSITIVE LOGITS

we

1.44

 they

1.32

We

1.28

We

1.24

They

1.14

1.13

 They

1.12

 THEY

1.12

THEY

1.08

we

1.07

Activations Density 0.150%

the usage of the word "we" and variations of it to highlight collective identity or group involvement

No Comments

No Known Activations