INDEX

Explanations

references to geographical locations and entities within a broader context

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

 betweenstory

-0.76

expandindo

-0.73

 ब्रेकडाउन

-0.66

MemoryWarning

-0.64

 GenerationType

-0.63

 disambiguazione

-0.62

MessageTagHelper

-0.60

GOTREF

-0.59

Diweddarwch

-0.57

xase

-0.55

POSITIVE LOGITS

 cancer

0.31

 "..\..\

0.30

 gọn

0.30

נם

0.30

sul

0.30

anmar

0.29

 online

0.29

tams

0.29

jublj

0.29

//

0.29

Activations Density 0.215%

references to geographical locations and entities within a broader context

No Comments

No Known Activations