INDEX

Explanations

mathematical expressions and numerical values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

edo

-0.52

Vil

-0.46

ματο

-0.45

ise

-0.45

อบ

-0.41

ydon

-0.41

rup

-0.41

 انف

-0.41

ด

-0.41

나

-0.40

POSITIVE LOGITS

featureID

1.07

 AssemblyCulture

1.06

 unknownFields

1.06

DockStyle

1.05

oredCriteria

1.04

LookAnd

1.02

 betweenstory

0.98

MigrationBuilder

0.97

complexContent

0.96

 '\\;'

0.95

Activations Density 1.314%

mathematical expressions and numerical values

No Comments

No Known Activations