INDEX

Explanations

instances of the word "to" indicating intentions or purposes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.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

Personensuche

-0.98

IndentedString

-0.86

ⓧ

-0.80

NameInMap

-0.71

 venait

-0.70

complexContent

-0.70

GEBURTSDATUM

-0.68

 Wiese

-0.67

 meint

-0.66

 realisation

-0.65

POSITIVE LOGITS

be

0.80

 have

0.79

UrlResolution

0.77

can

0.77

"):

0.77

)";

0.75

 dalamnya

0.75

 take

0.75

“

0.75

do

0.73

Activations Density 0.587%

instances of the word "to" indicating intentions or purposes

No Comments

No Known Activations