INDEX

Explanations

mentions of defense and military-related entities and contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_10/width_16k/average_l0_113

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

IsMutable

-0.71

PropertyChanging

-0.69

httphttps

-0.68

ſicht

-0.68

GEBURTSDATUM

-0.68

RegressionTest

-0.66

LookAnd

-0.65



-0.63

OGND

-0.63

expandindo

-0.63

POSITIVE LOGITS

 military

0.69

military

0.61

 Pentagon

0.57

 Military

0.56

Military

0.56

 militares

0.52

 militer

0.52

 MILITARY

0.51

 militaire

0.50

 militar

0.49

Activations Density 0.340%

mentions of defense and military-related entities and contexts

No Comments

No Known Activations