INDEX

Explanations

terms related to popular culture, specifically comics, movies, and animated series

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_23/width_131k/average_l0_101

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.23.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

ंदीखरीदारी

-0.60

IndentedString

-0.54

WithIOException

-0.53

 prestazioni

-0.49

 informée

-0.49

 automatiques

-0.47

ErrUnexpectedEOF

-0.46

 zdrav

-0.46

RTDA

-0.46

/**

-0.45

POSITIVE LOGITS

 themed

0.56

themed

0.50

 темати

0.49

 themes

0.49

BeginInit

0.44

THEME

0.43

 theme

0.43

 lenker

0.43

tagHelperRunner

0.43

 lovers

0.43

Activations Density 0.028%

terms related to popular culture, specifically comics, movies, and animated series

No Comments

No Known Activations