INDEX

Explanations

numerical values and their mathematical relationships

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

 فريبيس

-0.64

 Lint

-0.51

醐

-0.51

 réfléchir

-0.49

 pisau

-0.48

 Panel

-0.47

ascin

-0.47

 hear

-0.47

wyżs

-0.46

 prefent

-0.45

POSITIVE LOGITS

SourceChecksum

0.76

Tikang

0.70

 quarter

0.66

NameInMap

0.63

 kasarigan

0.57

UnsafeEnabled

0.55

WebServlet

0.55

 getItemId

0.54

はじめに

0.54

 BorderRadius

0.53

Activations Density 0.555%

numerical values and their mathematical relationships

No Comments

No Known Activations