INDEX

Explanations

references to human-centric concepts and rights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 aniversario

-0.73

 Trotz

-0.68

ValueStyle

-0.68

 Diwali

-0.67

łość

-0.65

 aikana

-0.64

 productivo

-0.64

 isles

-0.63

 Nomenclature

-0.62

kulum

-0.62

POSITIVE LOGITS

 human

2.47

human

2.21

 Human

2.19

 HUMAN

2.17

Human

2.15

HUMAN

2.06

 humans

1.93

 Humans

1.74

 humano

1.72

 humanos

1.69

Activations Density 0.084%

references to human-centric concepts and rights

No Comments

No Known Activations