INDEX

Explanations

references to numerical data and coding syntax

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

code values sequence length vary

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 disambiguazione

-0.56

 <=",

-0.49

WriteBarrier

-0.47

клопе

-0.44

writeFieldEnd

-0.42

 kaynağından

-0.41

 Buchung

-0.41

majánló

-0.40

 Italijani

-0.40

 tempio

-0.40

POSITIVE LOGITS

IsPostBack

0.47

出版年

0.41

WebElementEntity

0.38

 onPostExecute

0.38

saraba

0.37

 Olímpicos

0.37

slidesToShow

0.37

CloseOperation

0.36

lacable

0.35

 invokingState

0.35

Activations Density 0.283%

references to numerical data and coding syntax

code values sequence length vary

No Comments

No Known Activations