INDEX

Explanations

references to American nationality or identity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_mlp_out

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

 American

-2.50

American

-2.30

 AMERICAN

-1.96

 american

-1.92

american

-1.90

AMERICAN

-1.80

 Americans

-1.52

Americans

-1.32

 America

-1.30

 Amerikan

-1.27

POSITIVE LOGITS

PerformLayout

0.74

 الحره

0.73

Hochspringen

0.71

 AssemblyCulture

0.66

MessageTagHelper

0.66

AutoresizingMask

0.61

 BrowserModule

0.60

queryInterface

0.60

makeText

0.60

ことができました

0.59

Activations Density 0.125%

references to American nationality or identity

No Comments

No Known Activations