INDEX

Explanations

references to the USA, particularly in news contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 3-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_3/width_16k/average_l0_59

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.3.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

...

-0.72

 potential

-0.68

-0.62

ch

-0.61

\_

-0.60

ه‌

-0.60

 XCTest

-0.59

...

-0.59

getCurrent

-0.59

POSITIVE LOGITS

USA

2.46

USA

2.39

usa

1.51

Usa

1.50

usa

1.36

Usa

1.25

 متعلقه

0.95

 USAID

0.90

 للاسماء

0.87

ExecuteReader

0.83

Activations Density 0.066%

references to the USA, particularly in news contexts

No Comments

No Known Activations