INDEX

Explanations

lists of items or structured data

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

lists of numbers

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

ReusableCell

-0.51

httphttps

-0.51

Bakgrunnsstoff

-0.48

 GenerationType

-0.48

contentLoaded

-0.48

HtmlAttribute

-0.46

 AssemblyCulture

-0.46

spesies

-0.45

PRNewswire

-0.45

 Roskov

-0.45

POSITIVE LOGITS

MigrationBuilder

0.40

而

0.37

{?}

0.35

BnF

0.35

izr

0.34

 draft

0.33

glPopMatrix

0.32

 Rake

0.31

draft

0.31

/>

0.31

Activations Density 0.042%

lists of items or structured data

lists of numbers

No Comments

No Known Activations