INDEX

Explanations

words and phrases referencing philosophical concepts and existential questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 33-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_16k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.33.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

WebElementEntity

-0.57

 StatelessWidget

-0.42

 StatefulWidget

-0.40

höhung

-0.37

Normdaten

-0.37

+#+#

-0.37

 pysty

-0.36

Datuak

-0.36

 learns

-0.36

StrictMode

-0.35

POSITIVE LOGITS

 excites

1.23

 fascin

1.17

 aston

1.16

 amaze

1.10

 frighten

1.10

 annoys

1.09

 pleases

1.09

 impress

1.09

 excite

1.09

 scares

1.05

Activations Density 0.682%

words and phrases referencing philosophical concepts and existential questions

No Comments

No Known Activations