INDEX

Explanations

references to surveillance or governmental oversight

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-mlp-65k

Configuration

google/gemma-scope-2b-pt-mlp/layer_18/width_65k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.18.hook_mlp_out

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

extAlignment

-0.81

ietal

-0.57

IonicModule

-0.57

MENAFN

-0.56

ReactDOM

-0.51

rungsseite

-0.50

 McKinsey

-0.49

ExtendWith

-0.48

örös

-0.48

UnknownFields

-0.47

POSITIVE LOGITS

Skocz

0.61

parsedMessage

0.60

 Flickr

0.57

CppMethod

0.57

vaar

0.56

 MyBatis

0.56

NameInMap

0.52

>{@

0.52

Personensuche

0.51

 JpaRepository

0.51

Activations Density 0.243%

references to surveillance or governmental oversight

No Comments

No Known Activations