INDEX

Explanations

terms related to intelligence and reconnaissance

oai_token-act-pair · gpt-4o-mini

New Auto-Interp

Configuration

chanind/gemma-2-2b-batch-topk-matryoshka-saes-w-32k-l0-40/standard/blocks.12.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

32,768

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Architecture

jumprelu

Context Size

1,024

Dataset

chanind/pile-uncopyrighted-gemma-1024-abbrv-1B

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

@"/

-0.68

 indisponible

-0.63

Precautionary

-0.63

uxxxx

-0.63

 Ivor

-0.62

henko

-0.61

oriasis

-0.59

gorm

-0.59

 Firestone

-0.57

Subject

-0.56

POSITIVE LOGITS

 Intelligence

1.20

 intelligence

1.16

spy

1.16

Spy

1.15

 espionage

1.12

Intelligence

1.10

intelligence

1.08

 spies

1.08

CIA

1.01

 spying

1.00

Activations Density 0.016%

terms related to intelligence and reconnaissance

No Comments

No Known Activations