INDEX

Explanations

recommendations for reading or watching content

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

followed by prepositions/conjunctions

np_acts-logits-general · gemini-2.0-flash

engage with content

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 28-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_28/width_16k/average_l0_119

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.28.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

 nakalista

-0.45

LookAnd

-0.43

 Meksiku

-0.41

HideFlags

-0.38

Anhalt

-0.38

 елның

-0.38

 vVar

-0.37

ɵɵ

-0.36

yelitis

-0.36

sou

-0.35

POSITIVE LOGITS

AndEndTag

0.56

 EconPapers

0.52

tagHelperRunner

0.50

šanai

0.50

 propOrder

0.48

 autón

0.46

 présidenti

0.45

 fotográfica

0.44

 invokingState

0.44

able

0.42

Activations Density 0.492%

recommendations for reading or watching content

followed by prepositions/conjunctions

engage with content

No Comments

No Known Activations