INDEX

Explanations

structured format indicators often used in data entries or code

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 6-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_6/width_16k/average_l0_133

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.6.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

Datuak

-0.73

ⓧ

-0.68

oa̍t

-0.66

Хьажоргаш

-0.59

NewUrlParser

-0.55

كتشاف

-0.53

querade

-0.51

})$}

-0.50

ascin

-0.50

<tbody>

-0.50

POSITIVE LOGITS

barry

0.52

hane

0.51

She

0.51

postmedia

0.51

 vaikka

0.50

 laundering

0.47

 esposa

0.44

ScopeManager

0.44

 dibu

0.43

stu

0.43

Activations Density 0.106%

structured format indicators often used in data entries or code

No Comments

No Known Activations