INDEX

Explanations

structured formats used in interviews and research studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

ContentLoaded

-0.64

 EconPapers

-0.58

GEBURTSDATUM

-0.55

 šķ

-0.54

 CommonModule

-0.51

SOUNDBITE

-0.50

ủng

-0.50

 ویکی‌پدیای

-0.49

 Paglinawan

-0.48

Edg

-0.48

POSITIVE LOGITS

 individually

1.03

 individual

0.97

 individualized

0.92

OneToOne

0.89

one

0.87

individual

0.87

 personalized

0.86

one

0.84

Individual

0.83

一对

0.83

Activations Density 0.164%

structured formats used in interviews and research studies

No Comments

No Known Activations