INDEX

Explanations

phrases related to visual perception and attention

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

hdessä

-0.55

 viewDidLoad

-0.54

.*")]

-0.51

ponses

-0.51

enerys

-0.50

StreetMap

-0.49

 BoxDecoration

-0.48

complexType

-0.48

 ویکی‌پدیای

-0.47

 høj

-0.47

POSITIVE LOGITS

 eyes

2.55

eye

2.49

Eye

2.07

 Eyes

2.07

eye

2.04

Eye

2.01

eyes

1.96

 EYES

1.95

EYE

1.89

Eyes

1.88

Activations Density 0.155%

phrases related to visual perception and attention

No Comments

No Known Activations