INDEX

Explanations

references to reading and personal opinions about books

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_1/width_16k/average_l0_69

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

󠁢

-0.60

 للمعارف

-0.52

PhysRev

-0.50

ServletConfig

-0.45

脚注の使い方

-0.45

 disambiguazione

-0.45

rungsseite

-0.44

 Verſ

-0.42

 estekak

-0.42

ッポン

-0.41

POSITIVE LOGITS

 ModelExpression

0.58

awtextra

0.53

ready

0.52

 read

0.52

 Step

0.50

 ready

0.49

odd

0.49

 odds

0.49

Ready

0.48

Werdegang

0.45

Activations Density 0.192%

references to reading and personal opinions about books

No Comments

No Known Activations