INDEX

Explanations

statements asserting the truth or falsity of mathematical or logical propositions

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

GIVEREF

-0.56

IVEREF

-0.54

 snippetHide

-0.54

 Италијани

-0.52

uxxxx

-0.51

AndEndTag

-0.51

ंदीखरीदारी

-0.51

怎麼辦

-0.48

 Gebet

-0.48

 Bewußt

-0.48

POSITIVE LOGITS

 false

0.65

 False

0.64

false

0.59

FALSE

0.57

False

0.54

 FALSE

0.54

 True

0.49

 true

0.47

 statements

0.45

 untrue

0.45

Activations Density 0.067%

statements asserting the truth or falsity of mathematical or logical propositions

No Comments

No Known Activations