INDEX

Explanations

references to damage, injuries, or harm, and associated legal contexts or awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

BagLayout

-0.54

ValueStyle

-0.49

 Voltaire

-0.47

setVerticalGroup

-0.45

Parcelize

-0.44

 impl

-0.44

 Majefty

-0.44

Réalisation

-0.43

iNdEx

-0.43

HomeAsUpEnabled

-0.43

POSITIVE LOGITS

 award

0.75

 destrucción

0.70

 destruction

0.66

Award

0.64

 Award

0.63

 awards

0.63

 premios

0.62

award

0.61

Known

0.59

 damage

0.58

Activations Density 0.280%

references to damage, injuries, or harm, and associated legal contexts or awards

No Comments

No Known Activations