INDEX

Explanations

phrases related to conflict or warfare

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

SourceChecksum

-0.74

abetes

-0.67

NameInMap

-0.61

+};

-0.60

HtmlAttribute

-0.57

abestanden

-0.55

Composable

-0.55

shutil

-0.53

 joaat

-0.52

thebibliography

-0.52

POSITIVE LOGITS

 attack

2.03

 attacks

1.86

attack

1.76

 attacked

1.65

 ATTACK

1.63

 Attack

1.61

 attacking

1.60

 Attacks

1.54

attacks

1.47

Attack

1.45

Activations Density 0.214%

phrases related to conflict or warfare

No Comments

No Known Activations