INDEX

Explanations

sections of text that list or describe acronyms and abbreviations

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

pse

-0.48

 незавершена

-0.47

Geplaatst

-0.47

setopt

-0.46

󠁣

-0.45

dflare

-0.44

wapV

-0.43

 Centa

-0.43

EnableWeb

-0.43

Composable

-0.42

POSITIVE LOGITS

 ProtoMessage

0.58

Personensuche

0.48

 esternos

0.48

StoreMessageInfo

0.46

 abbreviations

0.44

WebElementEntity

0.44

 resourceCulture

0.44

 symbols

0.42

 symboles

0.41

ChildScrollView

0.38

Activations Density 0.003%

sections of text that list or describe acronyms and abbreviations

No Comments

No Known Activations