INDEX

Explanations

bullet points or list items

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

ILayout

-0.63

Tikang

-0.61

isNullOrEmpty

-0.58

ARANCE

-0.58

 onix

-0.58

gangssprach

-0.57

henswürdigkeiten

-0.57

lería

-0.57

tronlink

-0.56

 panier

-0.56

POSITIVE LOGITS

•

1.15

••

1.08

•••

1.00

••••

0.94

.•

0.86

•

0.86

 ••

0.81

~•

0.80

)•

0.78

°•

0.68

Activations Density 0.005%

bullet points or list items

No Comments

No Known Activations