INDEX

Explanations

phrases related to political interference and investigation

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-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

-0.68

-0.66

 very

-0.65

´

-0.62

del

-0.62

de

-0.60

↵

-0.59

di

-0.58

so

-0.58

POSITIVE LOGITS

 ainfi

1.50

 avoient

1.50

 Monfieur

1.46

 étoient

1.45

 enfans

1.45

 auroit

1.43

 plufieurs

1.39

 feroit

1.35

 pouvoit

1.26

 quæ

1.20

Activations Density 2.840%

phrases related to political interference and investigation

No Comments

No Known Activations