INDEX

Explanations

references to Stack Exchange platforms and related terms

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

CloseOperation

-0.85

 ***!

-0.74

 nahilalakip

-0.70

 Biôgrafia

-0.69

 ivelany

-0.68

OGND

-0.67

redients

-0.67

**/

-0.66

::_('

-0.65

-0.64

POSITIVE LOGITS

 stacks

1.21

 stack

1.14

 Stack

1.07

 Stacks

1.07

 STACK

0.98

Stack

0.95

STACK

0.89

stack

0.88

stacks

0.87

 pile

0.85

Activations Density 0.106%

references to Stack Exchange platforms and related terms

No Comments

No Known Activations