INDEX

Explanations

mentions of repetitive events or actions, particularly those indicating a return or recurrence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

Obrázky

-0.50

ogens

-0.48

WebVitals

-0.48

 AssemblyTitle

-0.48

 بيها

-0.46

FormTagHelper

-0.46

 Administrativna

-0.45

 pasto

-0.45

cerpts

-0.44

hibits

-0.44

POSITIVE LOGITS

 again

1.71

again

1.48

Again

1.45

 Again

1.43

 novamente

1.42

 erneut

1.41

 AGAIN

1.36

 nuevamente

1.25

 repeat

1.24

今回も

1.23

Activations Density 0.382%

mentions of repetitive events or actions, particularly those indicating a return or recurrence

No Comments

No Known Activations