INDEX

Explanations

definitions and clarifications of terms

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 surla

-0.52

Diweddarwch

-0.46

UnusedPrivate

-0.46

requestCode

-0.45

EndContext

-0.45

\{\\

-0.44

SequentialGroup

-0.43

 coö

-0.43

Наводи

-0.42

 Tenggara

-0.41

POSITIVE LOGITS

 AssemblyTitle

0.46

########.

0.43

 wrong

0.43

 incorrect

0.42

 правильно

0.39

 Administrativna

0.38

ͫ

0.38

Прави

0.36

tagPool

0.36

gifs

0.35

Activations Density 0.063%

definitions and clarifications of terms

No Comments

No Known Activations