INDEX

Explanations

repeated coding symbols or markers in a document

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

 ModelExpression

-0.77

 Chwiliwch

-0.67

CloseOperation

-0.66

 ویکی‌پدی

-0.64

 OMITBAD

-0.63

contentLoaded

-0.63

majánló

-0.62

 للمعارف

-0.59

addContainerGap

-0.59

LayoutStyle

-0.57

POSITIVE LOGITS

tgz

4.88

transQ

0.72

 betweenstory

0.69

fromnode

0.63

tar

0.61

tar

0.56

Tar

0.54

Tar

0.52

 ""],

0.52

TAR

0.50

Activations Density 0.000%

repeated coding symbols or markers in a document

No Comments

No Known Activations