INDEX

Explanations

references to weapons and military-related activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

ときの

-0.40

 pouvoirs

-0.40

 szeret

-0.40

skab

-0.39

lo

-0.39

aja

-0.38

nja

-0.38

 boucles

-0.38

헌

-0.38

atorship

-0.37

POSITIVE LOGITS

rungsseite

1.25

AddTagHelper

1.02

########.

0.95

 autorytatywna

0.93

ValueStyle

0.89

StoryboardSegue

0.85

Билгалдахарш

0.84

WireFormatLite

0.83

 kaynağından

0.83

Activations Density 0.150%

references to weapons and military-related activities

No Comments

No Known Activations