INDEX

Explanations

references to historical events or figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

ésultats

-0.76

 propOrder

-0.75

 للمعارف

-0.75

ðsíða

-0.74

 noDo

-0.73

$_(

-0.72

pulseira

-0.71

Personendaten

-0.71

rizona

-0.71

intenant

-0.70

POSITIVE LOGITS

par

0.31

li

0.28

 training

0.27

 regula

0.27

燐

0.26

med

0.25

 party

0.25

...

0.24

...

0.24

to

0.24

Activations Density 0.911%

references to historical events or figures

No Comments

No Known Activations