INDEX

Explanations

references to military units and ranks

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

 <>",

-0.56

RuleContext

-0.56

tagHelperRunner

-0.55

 autorytatywna

-0.54

TintMode

-0.46

 Biôgrafia

-0.46

kháu

-0.46

ùm

-0.45

<?

-0.44

يكب

-0.43

POSITIVE LOGITS

 regiments

0.59

 soldier

0.57

 recruits

0.56

 recruit

0.55

 recruitment

0.54

 army

0.54

soldier

0.53

 recruited

0.50

 disbanded

0.50

 regiment

0.50

Activations Density 0.504%

references to military units and ranks

No Comments

No Known Activations