INDEX

Explanations

phrases related to announcements and new plans

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 5-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_5/width_16k/average_l0_68

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.5.hook_resid_post

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

ibatis

-0.69

 getS

-0.60

'@/

-0.57

ledem

-0.56

Sar

-0.55

cuterie

-0.55

sø

-0.55

getFrom

-0.55

xue

-0.54

oeuvre

-0.54

POSITIVE LOGITS

 announce

2.26

 announced

2.24

 announces

2.19

 announcing

2.17

 announcements

2.06

 announcement

2.06

 Announced

1.95

announced

1.94

 ANNOUNC

1.91

announce

1.85

Activations Density 0.066%

phrases related to announcements and new plans

No Comments

No Known Activations