INDEX

Explanations

programming constructs and syntax elements related to functions and operations

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

code structures

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

 depended

-0.40

 estekak

-0.38

 generating

-0.38

RTEX

-0.38

ölkerung

-0.38

\{\\

-0.37

niająca

-0.36

 Umfang

-0.36

omel

-0.36

culature

-0.36

POSITIVE LOGITS

expandindo

0.80

 Biôgrafia

0.73

exitRule

0.68

 HttpNotFound

0.64

writeFieldEnd

0.60

PreferredItem

0.53

ActionCreators

0.50

 CanadaChoose

0.50

HandlerContext

0.50

AsUp

0.48

Activations Density 0.152%

programming constructs and syntax elements related to functions and operations

code structures

No Comments

No Known Activations