INDEX

Explanations

mathematical and programming terminology

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

greatest common divisor

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

}{*}{

-0.46

TypedDataSet

-0.43

舺

-0.43

Cid

-0.42

tonode

-0.42

WaitGroup

-0.41

 AssemblyCompany

-0.41

Cita

-0.41

BagLayout

-0.40

 ankara

-0.40

POSITIVE LOGITS

Derbyniad

0.68

 للاسماء

0.52

 تضيفلها

0.52

Jeografia

0.51

URLException

0.45

ьаж

0.39

UnknownFieldSet

0.37

 niyang

0.37

protoc

0.36

 consumer

0.35

Activations Density 1.191%

mathematical and programming terminology

greatest common divisor

No Comments

No Known Activations