INDEX

Explanations

specifications related to image dimensions and layout

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

!("{}",

-0.41

Herzliche

-0.37

serviceWorker

-0.37

BASEPATH

-0.36

URIComponent

-0.35

 still

-0.34

ware

-0.34

település

-0.34

mpä

-0.34

WebVitals

-0.33

POSITIVE LOGITS

 EconPapers

0.94

sidemargin

0.64

XmlAccessorType

0.60

 width

0.54

DockStyle

0.54

 GenerationType

0.52

 Paglinawan

0.52

 estekak

0.52

 maxWidth

0.51

IntoConstraints

0.51

Activations Density 0.009%

specifications related to image dimensions and layout

No Comments

No Known Activations