INDEX

Explanations

words and phrases related to health issues, behaviors, and wellness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

zyż

-0.57

Dichloroethene

-0.55

 Rape

-0.54

 poffe

-0.51

PLICIT

-0.51

 rape

-0.51

lē

-0.50

impresa

-0.50

 Taliban

-0.50

 felony

-0.49

POSITIVE LOGITS

 health

1.17

health

1.00

 healthier

0.87

 HEALTH

0.87

 saúde

0.86

HEALTH

0.86

 Health

0.85

 healthiest

0.83

Health

0.81

 healthy

0.80

Activations Density 0.664%

words and phrases related to health issues, behaviors, and wellness

No Comments

No Known Activations