INDEX

Explanations

references to significant 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 @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_0/width_16k/average_l0_105

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 препратки

-1.04

astify

-0.90

 تضيفلها

-0.87

 EconPapers

-0.87

 الاطلاع

-0.86

 nahilalakip

-0.84

InjectAttribute

-0.84

 kaarangay

-0.83

StructEnd

-0.82

consulté

-0.81

POSITIVE LOGITS

webElementXpaths

0.55

at

0.50

and

0.46

 meines

0.45

of

0.43

ConstraintMaker

0.42

etsu

0.42

ArrowToggle

0.42

Ihr

0.41

in

0.40

Activations Density 0.545%

references to significant numerical values

No Comments

No Known Activations