INDEX

Explanations

references to specific weeks and updates related to those weeks

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

 Glaser

-0.36

 Gentry

-0.33

 into

-0.32

 onto

-0.31

 Southern

-0.30

 Gilmour

-0.28

 egyéb

-0.28

 Brandt

-0.28

."

-0.28

 Dougherty

-0.28

POSITIVE LOGITS

Week

1.57

 week

1.45

week

1.41

WEEK

1.41

 Week

1.39

 WEEK

1.38

weeks

1.34

 weeks

1.25

 semana

1.22

 semaine

1.20

Activations Density 0.100%

references to specific weeks and updates related to those weeks

No Comments

No Known Activations