INDEX

Explanations

references to death or the deceased

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

 Macbeth

-0.67

BeginContext

-0.63

 hâte

-0.59

 meille

-0.56

 fratelli

-0.56

 veloce

-0.54

 hermanos

-0.53

 gusto

-0.52

롭

-0.52

kkor

-0.52

POSITIVE LOGITS

 dead

4.39

dead

3.75

 Dead

3.67

Dead

3.60

 DEAD

3.25

DEAD

2.66

 мерт

1.91

 muerto

1.78

 muertos

1.71

 lifeless

1.59

Activations Density 0.094%

references to death or the deceased

No Comments

No Known Activations