INDEX

Explanations

code components related to class definitions and data field declarations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

web

-0.48

and

-0.47

bar

-0.46

or

-0.46

poli

-0.45

os

-0.45

 back

-0.44

an

-0.44

POSITIVE LOGITS

 OnInit

2.10

ArgsConstructor

2.04

:✨

1.70

TagMode

1.68

 StatefulWidget

1.58

 StatelessWidget

1.58

SourceChecksum

1.45

UnusedPrivate

1.30

 linkovi

1.25

AndEndTag

1.22

Activations Density 0.006%

code components related to class definitions and data field declarations

No Comments

No Known Activations