INDEX

Explanations

references to notes or annotations

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

expandindo

-0.68

SuspendLayout

-0.57

parsedMessage

-0.56

 Waray

-0.55

 تضيفلها

-0.55

complexContent

-0.55

InitVars

-0.54

豋

-0.54

__(/*!

-0.51

WindowConstants

-0.50

POSITIVE LOGITS

Personensuche

0.49

IsMutable

0.38

навли

0.36

ApiModelProperty

0.36

 remember

0.35

setViewportView

0.35

ajur

0.34

MessageOf

0.34

Remember

0.33

Autowired

0.32

Activations Density 0.004%

references to notes or annotations

No Comments

No Known Activations