INDEX

Explanations

expressions of personal opinions and preferences

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

ServletConfig

-0.35

NameInMap

-0.34

AndEndTag

-0.33

mbic

-0.32

:✨

-0.31

PN

-0.30

gas

-0.29

 Trains

-0.29

 okuyayım

-0.29

 рее

-0.29

POSITIVE LOGITS

 betweenstory

0.67

endpush

0.57

 للمعارف

0.56

 isolato

0.54

ItemBackground

0.51

ScopeManager

0.50

 snippetHide

0.50

setVerticalGroup

0.48

ChildScrollView

0.48

oredCriteria

0.48

Activations Density 0.047%

expressions of personal opinions and preferences

No Comments

No Known Activations