INDEX

Explanations

numerical values or codes, likely related to software or coding contexts

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

IsContent

-0.50

simpleType

-0.44

 Italijani

-0.43

 onOptions

-0.43

encodeWith

-0.41

forName

-0.40

 jLabel

-0.40

AnchorStyles

-0.40

CloseOperation

-0.40

WithMany

-0.39

POSITIVE LOGITS

 snippetHide

0.59

criture

0.48

entibus

0.47

ftances

0.46

tagHelperRunner

0.46

ffions

0.46

Bureau

0.44

clerView

0.43

 Bureau

0.42

 rhestr

0.42

Activations Density 0.009%

numerical values or codes, likely related to software or coding contexts

No Comments

No Known Activations